ACS Publications. Most Trusted. Most Cited. Most Read
Using Machine Learning To Predict Suitable Conditions for Organic Reactions
More

OR SEARCH CITATIONS

My Activity
Publications

Figure 1Loading Img
Download Hi-Res ImageDownload to MS-PowerPointCite This:ACS Cent. Sci. 2018, 4, 11, 1465-1476
  • Open Access
Research Article

Using Machine Learning To Predict Suitable Conditions for Organic Reactions
Click to copy article linkArticle link copied!

Open PDFSupporting Information (1)

ACS Central Science

Cite this: ACS Cent. Sci. 2018, 4, 11, 1465–1476
Click to copy citationCitation copied!
https://doi.org/10.1021/acscentsci.8b00357
Published November 16, 2018

Copyright © 2018 American Chemical Society. This publication is licensed under these Terms of Use.

Request reuse permissions

Abstract

Click to copy section linkSection link copied!
High Resolution Image
Download MS PowerPoint Slide

Reaction condition recommendation is an essential element for the realization of computer-assisted synthetic planning. Accurate suggestions of reaction conditions are required for experimental validation and can have a significant effect on the success or failure of an attempted transformation. However, de novo condition recommendation remains a challenging and under-explored problem and relies heavily on chemists’ knowledge and experience. In this work, we develop a neural-network model to predict the chemical context (catalyst(s), solvent(s), reagent(s)), as well as the temperature most suitable for any particular organic reaction. Trained on ∼10 million examples from Reaxys, the model is able to propose conditions where a close match to the recorded catalyst, solvent, and reagent is found within the top-10 predictions 69.6% of the time, with top-10 accuracies for individual species reaching 80–90%. Temperature is accurately predicted within ±20 °C from the recorded temperature in 60–70% of test cases, with higher accuracy for cases with correct chemical context predictions. The utility of the model is illustrated through several examples spanning a range of common reaction classes. We also demonstrate that the model implicitly learns a continuous numerical embedding of solvent and reagent species that captures their functional similarity.

Copyright © 2018 American Chemical Society

Subjects

what are subjects

Synopsis

Machine learning model predicts conditions for organic synthesis reactions and quantifies solvent and reagent similarity.

Introduction

Click to copy section linkSection link copied!
Retrosynthetic planning is the process of proposing pathways to synthesize target molecules from available starting chemicals, and has demonstrated its importance and success in the chemical industry. (1,2) While retrosynthesis traditionally requires extensive training and expertise of a chemist, recent years have seen renewed interest in computer-assisted synthetic planning (CASP). (3−7) With the application of powerful machine learning techniques to large data sets of organic reactions like Reaxys (8) and the USPTO database, (9) there have been major advances both in searching for possible retrosynthetic pathways (10−16) and in evaluating the feasibility of the proposed reactions. (5,17−22)
While existing tools have been demonstrated to predict the likelihood of success of reactions with good accuracy, (20) one obstacle to experimentally validating computer-proposed reactions is the specification of reaction conditions, including chemical context (catalyst, reagent, solvent) and other operating conditions (e.g., temperature, pressure). In some cases, small changes in reaction conditions can lead to drastically different reaction outcomes. Therefore, recent work on reaction outcome prediction has started to include reaction conditions to improve the accuracy and specificity of predictions. (20,21) More importantly, reaction conditions are necessary to evaluate opportunities for one-pot synthesis, telescoping in flow, and amenability to the use of “green” solvents for sustainability.
Extensive work has been done on the optimization of conditions for specific reaction classes, using a combination of domain knowledge and empirical optimization techniques to automatically identify the best reaction condition. (23−27) Nevertheless, the initial guess of reaction conditions for a new reaction is predominantly considered a human task. Chemists use heuristics and perceived similarity of new reactions to ones they are familiar with to propose candidate conditions. However, this approach has its limitations and challenges. The recommendation might be biased by chemists’ preference and familiarity with certain types of reactions; the heuristic rules might not be all-encompassing or too abstract to narrow down to specific chemicals, and conditions of a precedent reaction may not be applicable to the new reactions even if the reactants are structurally similar.
In addition to the potential utility of in silico condition recommendation tools to practicing synthetic chemists, they are a necessary component of computer-aided synthesis planning. With thousands of plausible reactions generated in a few minutes or even seconds, it would be impossible to rely on manual input for suggesting reaction conditions. Research has shown that failure to specify appropriate reaction conditions might result in false prediction of reaction outcomes. (21)
However, computational condition recommendation remains a rarely addressed and complex challenge. Most existing work focuses only on specific elements of the chemical context (e.g., only reagents or only solvents), or specific reaction classes. Solvent selection, for instance, has been widely studied as a standalone problem. (28) Struebing et al. combined quantum mechanical (QM) calculations with a computer-aided molecular design procedure to identify solvents that accelerate reaction kinetics. (29) This approach was demonstrated to be effective for specific examples, yet it is difficult to apply at a larger scale due to the high computational cost of QM calculations.
Data-driven approaches have been employed to recommend conditions for specific types of reactions. Marcou et al. (30) built an expert system to predict the type of catalyst and solvent used for Michael additions, trained on 198 known reactions. The problem was formulated as multiple binary classification subproblems of whether a certain type of solvent/catalyst would be suitable for a specific Michael reaction. However, on an external test set, only 8 out of 52 reactions had both predicted solvent and catalyst matching the true context. Lin et al. used a similarity-based approach to recommend catalysts for desired deprotection reactions, and demonstrated the approach in catalytic hydrogenation reactions. (31)
A study by Segler and Waller tackles a broader scope of reactions using a knowledge graph model of organic chemistry to infer complementary and analogous reactivity. (32) Novel reactions are treated as missing links in this graph. Reaction context is taken as the combination of the first reactions that are linked with reactant molecules. They tested this approach on 11 reactions from the literature, and for most of them the model was able to identify the exact same or similar reagent/catalyst as used in the literature. This work demonstrated the feasibility of reaction context inference based on reaction patterns, yet context compatibility and temperature prediction are not taken into consideration.
Similar to the aforementioned approach, one straightforward method for identifying reasonable reaction conditions is to find a similar reaction in the literature and simply employ exactly the same reaction conditions reported for that precedent, referred to as the nearest-neighbor approach. Indeed this is an approach that many chemists may use implicitly. This can be successful with a database of known reactions that is sufficiently large and densely populated, but computationally, a nearest-neighbor search against millions of species is RAM- and CPU-intensive, even with optimized search strategies (e.g., using a ball tree). Furthermore, if some information in the nearest-neighbor reaction is not present (i.e., data is incomplete), that information cannot be inferred. The rigidity of this approach precludes asking questions essential to synthesis planning, such as whether the reaction could proceed in a particular replacement solvent.
In summary, we identify some primary limitations of existing approaches:
(1)

There has not been a published method that accurately predicts complete reaction conditions (catalysts, solvents, reagents, and temperature) suitable for use with a very large reaction corpus.

(2)

The compatibility and interdependence of chemical context and temperature are not taken into account in previous approaches.

(3)

No previous studies have performed quantitative evaluation of reaction condition predictions on a large-scale reaction data set. There are two major challenges which have impeded progress: (i) There is not a machine readable large data set available with catalysts/solvents/reagents classified into different types. (ii) For the similarity-based approaches it is difficult to quantitatively assess the level of “correctness” of conditions when comparing entire sets of conditions associated with different literature reactions.

(4)

Closer attention should be paid to balancing the generality/specificity of representing chemical context. If the representation is too general, such as manually encoded types/groups, it might not fully characterize functionality, and if it is too specific, e.g., copy–pasting the entire conditions from other reactions, it does not provide further information about chemical similarity.

New tools are needed that propose reaction conditions intelligently and can handle a broad scope of reaction classes. In this work we develop a neural-network-based model to predict suitable reaction conditions for arbitrary organic transformations. The model is trained on roughly 10 million examples from the Reaxys (8) database to predict the chemical species used as catalysts, solvents, reagents, and an appropriate temperature for the reaction. Prediction results are evaluated both quantitatively, using a variety of accuracy metrics, and qualitatively, using multiple sets of representative examples. It is also demonstrated that the model learns the similarity of the chemical context (e.g., different solvents/reagents) exclusively from reaction data.

Results and Discussion

Click to copy section linkSection link copied!
A neural network model is trained to predict up to one catalyst, two solvents,  two reagents, and the temperature for a given organic reaction. Detailed data processing and model formulation steps are described in the Methods section. The training process is essentially a multiobjective optimization that minimizes the overall loss function which is a weighted sum of the loss for each individual objective (namely, catalyst, solvent 1, solvent 2, reagent 1, reagent 2, and temperature). The progress of training is reflected in the change of the overall and individual loss, which is shown in Figure 1. Validation loss (dashed red line) decreases for 2 epochs and then reaches a plateau and stays higher than the training loss (solid red line). Based on the plotted losses in Figure 1 the first solvent (s1, yellow lines) and the first reagent (r1, orange lines) are the most difficult to predict, with a significantly higher loss value than the other objectives (not including temperature). There is a large fraction of reactions which do not have a second solvent (s2, blue lines) or reagent (r2, gray lines), in which cases the model only needs to predict the NULL class, making these second predictions easier to classify. The same principle applies to catalyst, where many reactions do not use a catalyst, and others have catalysts frequently recorded as reagents. The mean squared error for temperature (T loss, green lines) decreases steadily over the epochs, ending in 0.46 for the training set and 0.50 for the validation set (after scaling by a factor of 0.001 K−2).

Figure 1

Figure 1. Change of the loss functions with the number of epochs (left figure, overall; right figure, chemical context and temperature).

High Resolution Image
Download MS PowerPoint Slide

Statistical Analysis

Evaluating the results of chemical context prediction is a nontrivial task, mainly because it is a combination of individual chemicals, and because we lack a standard way of quantifying the “closeness” of the prediction when each exact chemical is not predicted. Since there is often more than one possible context combination suitable for a reaction, we do not want to focus exclusively on the top-one prediction, but also want to examine other highly ranked suggestions. However, the number of total combinations grows in a polynomial way (to the fifth power) with the increase of number of candidates to include for every individual element; e.g., if the top-three candidates are to be examined for the catalyst, both solvents, and both reagents, the total number of combinations is 243 (35), which is almost impossible to evaluate manually, and difficult to analyze. Since the data for catalyst, solvent 2 and reagent 2 are much more sparse than solvent 1 and reagent 1, there is likely more value in examining longer candidate list for the latter two. Therefore, we use the top-three reagent 1 predictions and top-three solvent 1 predictions along with the top-two catalyst, top-one solvent 2, and top-one reagent 2 to construct 18 top combinations, from which we can pick the top-three or top-10 combinations with the highest overall scores, calculated as the product of softmax probabilities for each individual element. The number of top candidates is a heuristic choice and can be tailored by model users for specific needs (e.g., the user can explore a longer list of catalyst candidates instead of only the top-two choices).
Evaluation is performed on the entire test set, and the accuracy values described in the Methods section are shown in Table 1. For the most difficult tasks of predicting solvent 1 and reagent 1, the frequency with which the recorded chemical  is found in the top-10 combinations is 83.0% and 83.1%, respectively. After including those close match predictions (defined in the Methods section), the accuracy for solvent 1 increases by a margin of 2.4%, and the accuracy for reagent 1 increases by 1.8%. This suggests that, though not explicitly coded, the model learns the chemical similarity of solvents/reagents/catalysts which tend to be used in closely related reactions.
Table 1. Accuracy of Prediction of the Chemical Context by the Condition Recommendation Modela
Prediction taskTop-3 exact matchesTop-10 exact matchesTop-3 close matchesTop-10 close matches
c93.6%94.9%94.9%96.4%
s175.8%83.0%78.2%85.4%
s290.1%91.7%90.2%91.9%
r173.2%83.1%74.8%84.9%
r289.3%91.8%89.3%92.1%
c, s1, r157.3%66.0%60.4%69.6%
c, s1, s2, r1, r250.1%57.3%53.2%60.3%
a

c, s1, s2, r1, and r2 refer to catalyst, solvent 1, solvent 2, reagent 1, and reagent 2, respectively.

Compared to the high accuracies for the individual prediction tasks, the top-three accuracy for the full condition recommendation (catalyst, two solvents, two reagents) is 50.1%, and the top-10 accuracy is 57.3% (53.2% and 60.3% when including close match predictions). However, given that these numbers represent the requirement to predict the full combination (all five elements) of the exact recorded context, it is expected to be more challenging than predicting individual elements. We further computed the top-10 accuracy of a subset of the combinations—catalyst, solvent 1, and reagent 1—as 66.0% and 69.6% for exact matches and close match predictions, respectively.
To evaluate the meaningfulness of the accuracies given in Table 1, we compared the trained model with a baseline model, where top-10 combinations are chosen based on the frequencies of the catalysts, solvents, and reagents (Supporting Information, Table S1). Detailed comparisons and statistical parameters are given in the Supporting Information, Table S2. It can be seen that 87.3% of the reactions do not use a catalyst; 85.6% of the reactions do not use a second solvent, and 82.3% of the reactions do not use a second reagent. For these tasks, predicting the NULL class can achieve relatively high accuracy, but the trained model still performs better by a significant margin. Meanwhile, the top-3 accuracy for predicting the correct combination of chemical context is only 4.7%, compared to 50.1% in the trained model, indicating that simply using the most frequent combination of chemical context is not an effective method.
The evaluation of temperature is less straightforward than it seems, because the prediction of temperature is dependent on the chemical context, which means, in the top-10 temperature predictions, at least nine of them are based on chemical contexts that are at least partially different from the recorded context. Practically speaking, however, temperature is relatively easy to change and test in experiments, so suggesting an approximate initial guess of the temperature would be sufficiently helpful for the setup of experiments. The top-one temperature prediction falls within the ±10 or ±20 °C range of the recorded temperature in 36.7% and 57.7% of test cases, respectively. If we isolate reactions whose predicted chemical context matches the recorded chemical context, these accuracies increase to 42.6% and 65.9%. The mean absolute error (MAE) of temperature prediction for all reactions in the data set is 25.5 °C, and when the correct chemical context is found the MAE is 19.4 °C. Figure 2 visualizes this by plotting the predicted temperature against the recorded temperature for a 1% sample of the testing set. The fact that the quality of temperature prediction is significantly improved with correctly predicted chemical context demonstrates that the prediction of temperature accounts for the compatibility with the chemical context. This performance is also compared with a baseline model that predicts the most frequent temperature (20 °C) for every reaction in the Supporting Information. Figure S4 shows the distribution of temperature for the test set. While a majority of reactions use 20 °C, the distribution spans a wide range, and simply predicting the room temperature (20 °C) will result in a mean absolute error of 35.3 °C, which is significantly larger than the prediction given by the trained model and would be misleading for reactions that require high or low temperatures.

Figure 2

Figure 2. Relationship between the true temperature and the top-one predicted temperature (left panel), and predicted temperature if the predicted context matches the chemical context (right panel).

High Resolution Image
Download MS PowerPoint Slide

Qualitative Evaluation of Reaction Examples

In addition to the statistical analyses, qualitative evaluation of reaction examples helps provide chemical insight in the model predictions. We select reactions from a variety of common types of organic reactions to evaluate the quality of model predictions. We randomly select example reactions that are labeled by reaction type (around five for each type) and compare the true condition, top-one prediction, and the closest prediction within top-10 candidates. The closest prediction is defined as the prediction that has the largest number of chemical elements exactly matching the true chemical context. The reaction types we choose to test include hydrolysis, esterification, alkylation, epoxidation, Wittig, reduction, oxidation, deprotection, Suzuki–Miyaura coupling, Grubbs metathesis, and Buchwald–Hartwig amination. Due to space limitation, we only place a small part of the examples here in the main text with the top-one prediction (Figure 3) and the rest in the Supporting Information (Table S3) with both the top-one prediction and the closest prediction.

Figure 3

Figure 3. Example of model predictions compared with recorded context (temperature rounded to the closest integer; black text represents the recorded conditions, and blue text represents the predicted conditions). (A) Nucleophilic epoxidation. (B) Deprotection of fluorenylmethyloxycarbonyl (Fmoc). (C) Luche reduction of eneone, TBS = tert-butyl(dimethyl)silyl. (D) Buchwald–Hartwig aryl amination, BINAP = 2,2′-bis(diphenylphosphino)-1,1′-binaphthyl. (E) Suzuki-Miyaura coupling, CyJohnPhos = (2-biphenyl)dicyclohexylphosphine. (F) Hoveyda–Grubbs cross metathesis.

High Resolution Image
Download MS PowerPoint Slide
Figure 3A is an epoxidation reaction where two C═C bonds are present. (33) The recorded reagents for a nucleophilic epoxidation, selective for the electron deficient alkene, are correctly predicted by the model. Figure 3B shows a Fmoc deprotection reaction where either acidic or basic conditions can be used. In this case, basic conditions are proposed by the model for removal of the Fmoc group which does not affect the other acetate protecting groups in the molecule. (34) Notably, the predicted reagent (piperidine) is highly similar structurally and functionally to the recorded reagent (morpholine), demonstrating the model’s ability to capture chemical similarity. Figure 3C is a Luche reduction that needs a Lewis acid catalyst to selectivity reduce the carbonyl, and the model recognizes this specificity by suggesting cerium(III). (35)Figure 3D is a Buchwald–Hartwig aryl amination that uses BINAP as the ligand. (36) The metal atoms, ligands, the base, and the solvents are all correctly predicted by the model. Figure 3E,F is a Suzuki–Miyaura coupling reaction (37) and Grubbs metathesis, (38) respectively, for which the model also successfully predicts the exact chemical context.
It is worth pointing out the success of predicting the context of catalytic reactions is partially attributable to a data cleaning step that moves transition metal compounds from the reagents field to the catalyst field. This only increases the overall accuracy by a small margin (∼0.5%), but it significantly improves the quality of model predictions for catalytic reactions in the case studies. For example, before data cleaning, Figure 3D had the catalyst predicted as a reagent, and Figure 3E missed the catalyst entirely. A complete list of 62 reactions from 11 reaction types can be found in the Supporting Information, Table S3. Finally, temperature is predicted within a reasonable error from the true temperature (Figure 3 and Supporting Information, Table S3), and examples with large deviations are mostly cases with chemical contexts predicted different from the recorded ones.
As mentioned in the Introduction, we also performed a nearest-neighbor search for these reaction examples, shown in the Supporting Information. The nearest-neighbor search is analogous to searching for similar reactions in reaction databases such as Reaxys or SciFinder. Results in the Supporting Information, Table S6, show that the nearest-neighbor method works comparably well in giving the first suggestion for many examples, but it is more than 10 000 times slower than the neural network method presented here. Therefore, it is prohibitively expensive to evaluate the nearest-neighbor approach on the entire test set, and only a qualitative comparison is given on the examples described above. This suggests that using the trained neural-network model can achieve significantly faster speed in finding reasonable condition recommendations compared to the nearest-neighbor search method, and thus can be better integrated with computer-assisted retrosynthetic programs.
The results in Figure 3 and the Supporting Information, Table S3, show selected results pulled out of the test set of some common classes of reactions such as oxidations, reductions, and coupling reactions. Although these are common reaction classes, there is much diversity in specific conditions to achieve the desired reactivity depending on the structure of the starting materials. The results in Figure 3 demonstrate that specific conditions can be chosen by the model for many reaction classes. We also include in the Supporting Information, Table S4, 100 completely random examples from the test set that will not have the restriction of the 11 reaction classes that we initially pulled examples from. It can be seen that the prediction performance is good for most of these random cases.
Nevertheless, it is important to also analyze examples with the most incorrect predictions. We calculate the sum of binary variables indicating whether the correct or similar c, s1, s2, r1, and r2 are predicted and use it as a crude measure of “correctness”. The test results are sorted in ascending order of this sum, and examples are drawn from the first 10 000 entries (this is about 1% of reactions from the test set with the quantitatively worst predictions) in the reaction types described above (five for each type, for some types fewer than five as the reactions with labeled reaction types are sparse in the database). A list of 40 reactions are generated and listed in the Supporting Information, Table S5 (part of the examples shown in Figure 4). Even in these ostensibly “worst” cases, many predictions are not unreasonable. Figure 4A represents a common type of data quality issue in the data set, where water is used to quench the reaction but recorded in the same way as other reagents, despite being incompatible and explosive with alkali metals in their pure form. (39) However, the model recommends reagents that are commonly used for a dissolving metal reduction reaction. Figure 4B represents another type of problem observed in the data set, where a multistage reaction is recorded as a single transformation. (40) In the original reference, the Grubbs reaction and reduction are two isolated steps, which is probably a rare case in the database, and the model fails to recognize these two reactions simultaneously. The top suggestion is indeed poor, whereas the 6th to 10th suggestions all recognize the Grubbs reaction and suggest a Grubbs catalyst (see the Supporting Information, Table S5). In Figure 4C, the model suggests using tetrakis palladium(0) where the true context is palladium(II) and triphenylphosphine which is commonly used to form palladium(0) in situ, so the prediction seems to be a viable context combination for the reaction. (41)Figure 4D is a case of an azide reduction reaction. (42) The model recognizes a reductant is needed, and suggests hydrogen as a first choice. The fourth suggestion of the model include two reagents: 1,3-propanedithiol and triethylamine (Supporting Information, Table S5), which is a more plausible alternative to the recorded reagent. More examples can be found in the Supporting Information, Table S5, and many of the “poorly matching” results are not close to the published procedures but are not altogether unreasonable suggestions.

Figure 4

Figure 4. Examples of the reactions with the fewest chemical elements matching the recorded context (temperature rounded to the nearest integer; black text represents the recorded conditions, and red text represents the predicted conditions). (A) Birch alkylation. (B) Hoveyda–Grubbs cross metathesis, TBS = tert-butyl(dimethyl)silyl. (C) Suzuki-Miyaura coupling. (D) Azide reduction.

High Resolution Image
Download MS PowerPoint Slide
In addition, we include a test of predicting context for Michael addition reactions that are the same as used by Marcou et al., (30) which shows significant improvement; results are presented in the Supporting Information.

Learned Embedding of Solvents and Reagents

While no explicit relationship between chemical species is included in the model, it implicitly learns the functional similarity of solvents and reagents through training, as it can suggest similar chemicals for the same reaction. Taking solvent as an example, the similarity information can be extracted from the neural network, specifically the weight matrix in the last hidden layer before the final likelihood prediction and softmax activation. If two rows in the weight matrix are similar, the model will tend to predict similar scores for the corresponding two solvents. In other words, each solvent can be represented by its corresponding row from the weight matrix. This representation contains information about how it is used in different reactions and can be used to characterize functional similarity, implicitly averaged over all training reactions. This is analogous to the word embedding in natural language modeling where discrete words are converted to vectors of real numbers which contain similarity information (word2vec), (43) though we arrive at the representation indirectly, so we call the vector “solvent embedding”.
To visualize the embedding of solvents, the top 50 solvents with the highest frequency in the data set are selected, and labeled manually into four types (nonpolar, polar nonprotic, protic, and halogenated). The embeddings of the solvents are normalized to account for their frequency of use and projected into a 2-D space using the t-distributed stochastic neighbor embedding (t-SNE) (44) technique, shown in Figure 5. It is worth noting that the solvent embedding vectors do not have a physical interpretation, and t-SNE is a technique that aims to preserve the similarity of the data points in a visualized low-dimensional space. Therefore, the axes on Figure 5 do not have direct meaningful representation. Nevertheless, it can be seen that the solvents of the same type are clustered in the same part of the graph, and we can even observe some chemically reasonable trend across clusters (e.g., increasing polarity from the bottom right to upper left). Some “close-neighbor” pairs agree well with chemical knowledge (e.g., benzene and toluene, methanol and ethanol). The model is not supplied with any information about polarity or other electronic properties, but it learns the difference by the ways that solvents are used in various reactions.

Figure 5

Figure 5. Embedding of the most common 50 solvents projected onto a two-dimensional space using t-SNE. Solvents are naturally clustered into their corresponding classes (manually annotated).

High Resolution Image
Download MS PowerPoint Slide
The same plot for the most common 50 reagent embedding is shown in Figure 6. The functionality of reagents is more diverse than solvents, and the label list includes inorganic acid, organic acid, Lewis acid, inorganic base, organic base, reductant, oxidant, and activating reagents. While a 2-D projection might be insufficient to fully represent the variance of functionality, it can be observed that reagents of similar function are reasonably clustered in the same area of the plot. Additionally, incompatible reagents are segregated from each other (e.g., acids and bases, reductants and oxidants) demonstrating the ability of the model to not only recommend similar reagents but also lower the chance of recommending reagents that will likely to be reactive with one another.

Figure 6

Figure 6. Embedding of the most common 50 reagents projected onto a two-dimensional space using t-SNE. Reagents are naturally clustered into their corresponding classes (manually annotated).

High Resolution Image
Download MS PowerPoint Slide
The solvent and reagent embeddings extracted from the condition recommendation model demonstrate the possibility of understanding chemical functionality through reaction data, and have many potential applications. Most directly, it can be used as a tool to identify the closest alternative for the currently used solvents/reagents that are, for example, less toxic, cheaper, etc. It can also be used as input features for other machine learning problems that are context dependent, e.g., evaluation of reaction outcomes and solvent/reagent property estimation.

Strengths and Limitations

The neural-network model developed in this work overcomes many of the challenges described in the Introduction as summarized below:
(1)

By training on ∼10 million reactions from Reaxys, the model covers a wide range of organic reactions.

(2)

With a hierarchical neural network, the model predicts all the elements in the reaction condition sequentially with interdependence and relatively high accuracy.

(3)

The chemicals are not precategorized into classes so predictions can point to the specific chemicals that might, for example, be used as either an acid or an oxidant based on the reaction (e.g., sulfuric acid). On the other hand, individual chemical species are modeled as separate entities so that functional similarity can be learned during training and extracted from the model.

(4)

The model recommends reaction conditions much faster (less than 100 ms for one reaction) than nearest-neighbor search methods, and allows quantitative evaluation of model predictions on a large scale. It can also be used for efficient condition recommendation for a large number of reactions suggested by computer-assisted retrosynthetic analysis. The reaction conditions can be utilized by forward evaluation tools to better predict reactivity, especially for condition dependent reactions. In addition, the learned embeddings of solvents and reagents can be used to quantify similarity of conditions of sequential reactions to estimate separation requirements, and to find potential green alternatives to toxic solvents/reagents, both of which are helpful for pathway-level route screening and prioritization.

Meanwhile, the current model also has some limitations, as summarized below:
(1)

Since the chemical context is predicted in a sequential manner, we must limit number of predictions at each stage to obtain approximate top-10 combinations in a short time period (similar to a beam search).

(2)

Truncating the data based on minimal frequencies of catalysts, reagents, and solvents lowers the total number of trainable parameters and avoids data sparsity issues during training, but also limits the ability to predict rare contexts that are used by highly specific reactions.

(3)

There are various other limitations imposed by the imperfection of the training data. For example, even after filtering, some reactions with multiple transformations remain which confuses model prediction, and the labeling of reagents is sometimes misleading (e.g., quenching chemicals included as a reagent); there are some duplicated records or different labels for the same chemical. While these situations are relatively uncommon in the entire data set, a better curated data set can potentially further improve the model performance.

Methods

Click to copy section linkSection link copied!

Overview

The task of condition prediction can be divided into two parts: chemical context prediction (catalysts, solvents, reagents) is treated as a set of multiclass classification problems (i.e., choosing chemical species from a fixed list), while temperature prediction is treated as a regression problem. Pressure is not considered in this scope because the majority of published reactions of interest are run at atmospheric pressure, and databases often do not record pressure for such cases. Concentration is also not considered because it is excluded from tabulated reaction databases, although sparse information is available in the form of mass or volume of some chemical species. As mentioned in the introduction, all these parts should be linked together in one model to solve these classification and regression problems in a hierarchical formulation to account for the interconnectedness of, e.g., solvent and catalyst selection. We choose to use neural networks to construct the model architecture, because of their flexibility to recognize highly nonlinear relationships and because the size and diversity of the data warrant a high-capacity model. The model is trained on published reaction data from Reaxys (8) to predict the recorded reaction conditions, after which we are able to infer suitable reaction conditions for novel reactions.

Data

We construct our training set starting from the Reaxys reaction database that consists of 53 million reaction records. The information we use includes the simplified molecular-input line-entry system (SMILES) strings of the reactants and products, the Reaxys chemical ID and SMILES (if available) or name for the catalysts, solvents and reagents, and the temperature for the reactions. Note that in this context reagents are explicitly distinct from reactants, with the former generally not contributing carbon atoms to the reaction (typical reagents include acids, bases, oxidants, reductants, etc.). We restrict our analysis to single-product and single-step reactions to better align with the application to computer-aided synthesis planning. In this context, single-product reactions are defined as Reaxys reaction entries with only one recorded major product. Few entries have multiple products (e.g., specifying all outcomes in cases of ambiguous site selectivity), only 6.3% in this data set. Reactions are preannotated to specify the number of reaction steps associated with the recorded transformation; this attribute is used to filter out all multistep reactions. Some reaction examples passing this filter could still be considered as requiring multiple steps (as shown in the Results section), but it is otherwise hard to systematically distinguish between single- and multi-step reactions purely based on reactant and product structures. Reactions without recorded structures (half reactions) or with structures that could not be parsed by RDKit (45) are removed. We also discard reaction entries with no reaction condition information (i.e., no catalyst, no solvent, no reagent, and no temperature), and restrict the maximum number of unique solvents and reagents per reaction to two per category, which will be further explained in the Model Structure section. We also note that there is some ambiguity between catalysts and reagents in Reaxys—many catalysts are recorded as reagents, causing the data to be sparser for catalysts and increasing the number of distinct reagents. This issue can hardly be eliminated completely since a strict separation between reagents and catalyst is difficult to achieve. As a data cleaning step to mitigate this issue, all chemicals that appear as catalysts or reagents that include transition metals in the name are marked as “catalyst exclusive” and consolidated to the catalyst field.
For the remaining 12.1 million reaction examples, we analyze the frequency of each solvent, reagent, and catalyst species. A minimum frequency filter is applied to remove rare catalysts, solvents, and reagents. Rare chemical species significantly increase the number of classes over which the condition prediction will be made, introducing concerns over data sparsity, and including them contributes little to model coverage (frequency vs rank plots for catalysts, reagents, and solvents are provided in the Supporting Information, Figures S1–S3). A minimum frequency of 100 is applied to solvents, reagents, and catalysts. The final number of classes—distinct chemical species—for solvents is 232, for reagents is 2247, and for catalysts is 803; this is an appreciable reduction from the original 11 246, 151 214, and 10 323, respectively. Meanwhile, the number of reactions filtered out by this criterion is only 676 848, around 5% of the total reactions. The number of reactions filtered out by each criterion is listed in Table 2.
Table 2. Number of Reactions by Each Filter Criterion
criterionnumber of reactions
originally from Reaxys53 143 003
temperature out of range56 235
multistep23 536 281
multiproduct3 335 439
missing SMILES (including half reactions)92 472
cannot be sanitized by RDKit1 693 625
no condition information9 684 738
exceeding one catalyst, two solvents, or two reagents2 645 058
using rare catalysts, solvents, or reagents676 848
final data set11 422 307
Temperature is treated as a continuous variable. For those reactions whose temperature is recorded as a range (e.g., 0–20 °C), the midpoint of the range is used. We require the reaction temperature to be within −100 and 500 °C, which comfortably contains the vast majority of organic transformations and only excludes chemistry outside the scope of this study (e.g., hydrocarbon cracking). After the full data preprocessing pipeline, there are a total of 11.4 M reaction records that result in the final data set.

Molecular Representation

Morgan circular fingerprints, as implemented in RDKit (with radius 2, calculated as bit vectors with length 16 384, stereochemistry information included, and feature-based invariants are not used), (45) are used to represent reactants and product species, as they are a commonly used descriptor of organic molecules. (11,15,46) Catalysts, solvents, and reagents are directly represented as one-hot vectors, with each different chemical species (more precisely, each chemical entity with a unique ID in Reaxys) representing a unique class. A NULL class is added for each element, to represent reactions where the corresponding element is not recorded (e.g., no reagent, etc.). The lack of a well-defined chemical structure for certain species (e.g., air) precludes a richer descriptor-based representation, and we find a one-hot representation to work well in practice.

Model Structure

The neural network takes the product fingerprint and reaction fingerprint as two inputs. A reaction fingerprint is calculated as the difference between product fingerprint and reactant fingerprint, which represents the substructures that change during the reaction. (46) Predictions are made sequentially so that information from precedent elements can be incorporated into the prediction of subsequent elements (e.g., the prediction of solvent will depend on what catalyst is chosen). Temperature is the final output of the model, such that it relies on the chemical context recommendations. The workflow of the model (shown graphically in Figure 7) is as follows:
(1)

Reaction and product fingerprints are concatenated and passed through two fully connected layers (ReLU activation, size 1000; ReLU activation, size 1000, with a 0.5 dropout) to generate a dense representation of the fingerprints (referred to as Dense FP).

(2)

Dense FP is passed through two fully connected layers (ReLU activation, size 300; Softmax activation, size 803) to predict the catalyst (or NULL) for the reaction.

(3)

The one-hot vector of the catalyst prediction is then concatenated with Dense FP and passed through two fully connected layers (ReLU activation, size 300; Softmax activation, size 232) to predict the first solvent (or NULL).

(4)

Step 3 is repeated for prediction of the second solvent (size 228), the first reagent (size 2240), and the second reagent (size 1979). The numbers are smaller than the total class of solvents/reagents because some solvents/reagents are only present in one of the fields (i.e., only as Solvent/Reagent 1 or Solvent/Reagent 2).

(5)

One-hot vectors of the catalyst, solvents, and reagents and Dense FP are all concatenated and passed through two fully connected layers (ReLU activation, size 300; Linear activation, size 1) to predict the temperature.

Figure 7

Figure 7. Graphical representation of the neural-network model for context recommendation (“Hard Selection” refers to setting the value of the maximal element to one and zero for the rest, although the output of each classification task is a probability distribution).

High Resolution Image
Download MS PowerPoint Slide
Notable features of the model construction are as follows:
(1)

One feature is the order of the prediction tasks. The earlier it appears in the model, the more that task is able to be performed solely based on the reaction, independent of the other predictions. We experimented with predicting single elements using fingerprint information only and found that the validation accuracy (top-one accuracy) is highest for catalyst (92.1%), and similar for solvent and reagent (60.6% and 60.6%, respectively). This is consistent with how chemists generally approach this problem manually, i.e., identify if the reaction requires a catalyst. Reagents are placed last in the sequential prediction, so that information about catalyst and solvent selection is included in predicting reagents, which have the most unique possibilities and a greater level of flexibility even when the catalysts and solvents are fixed.

(2)

Another feature is the number of catalysts, solvents, and reagents for each reaction. Most of the reactions in the data set have no catalyst or at most one catalyst recorded, so the number of catalysts is limited to one. A majority of reactions use one solvent, but there are still many examples that use multiple solvent or multiple reagent combinations. Few reactions in the data set use three or more solvents or reagents, so limiting the number of solvents and reagents to two for each category keeps the model in a reasonable size. The final model has 38 M parameters.

Training and Evaluation

The data set is split randomly into training/validation/test sets with a ratio of 80/10/10. The time-split strategy is not used because, in practice, the model for condition recommendation is likely to be mainly used for interpolating and generalizing historical condition information to new substrates, but not designing fundamentally novel conditions that have not been previously discovered. It is worth noting that there can be multiple records for the same reaction, possibly happening under different conditions. When multiple records of the same reaction exist (i.e., if multiple reactions share the same reactants and products, which is quite common), they are grouped during shuffling to guarantee that the reactions in the test set are not present in the training set. Our model has intermediate outputs (catalysts, solvents, and reagents) that are used as input for the next prediction, which resembles the features of a recurrent neural network, so we apply the teacher forcing technique (47) during training. It takes the ground truth output, instead of the predicted ones, as the input for the next prediction task. This technique has been shown to increase stability and accuracy of the training. Categorical cross-entropy is used as the loss function for the classification problems (i.e., catalysts, solvents, and reagents), and mean squared error is used for regression (i.e., temperature). A weighting factor of 0.001 is applied to temperature so that the numeric values of the loss functions are approximately on the same scale. Training continues until the validation loss does not improve over five epochs.
Evaluation is performed both quantitatively and qualitatively. We calculated the accuracy of the true combined chemical context as well as individual elements to be within the top-three and top-10 predicted combinations. Additionally, we extend this accuracy calculation to include some “close match” predictions. The similarity of solvents is characterized by the Euclidean distance between the Abraham parameters (28) of the two solvents if the parameters are available, and otherwise only exact matches are considered. The methanol–ethanol pair is used as a threshold to identify solvents that are close matches. Catalysts and reagents are classified as close matches if they have the same metal atoms (for organometallic compounds) or if their feature-based Morgan fingerprints are exactly the same. The feature definitions are as implemented in RDKit, adapted from the definitions in Gobbi et al., (48) which define some invariants that share the same feature, such as “halogen” (e.g., −Cl and −Br), “hydrogen bond donor” (−SH and −OH), and unusual atoms (not H, C, N, O, F, S, Cl, Br, I; e.g., Na+ and K+). A complete table of feature definitions is included in the Supporting Information (Table S8).
For temperature, we calculated the percentage of cases when the temperature is predicted to be within ±10 and ±20 °C of the recorded temperature.
Besides the quantitative analysis, examples were chosen from common types of chemical reactions to demonstrate the wide applicability of the model and provide more insights into the model prediction characteristics. Successful and unsuccessful predictions in the test set are presented and analyzed to demonstrate the performance of the model. In comparison, the performance of the nearest-neighbor model on these reactions is also tested, and results are discussed in the Supporting Information.
This is a computational study, and we do not expect high safety hazards to be encountered.

Conclusion

Click to copy section linkSection link copied!
A neural network model was developed for the task of reaction condition recommendation. Using a hierarchical design and training on about 10 million reactions from Reaxys, the model gives recommendations on the catalyst, solvent, reagent, and temperature to be used for any organic reaction. The model is tested on 1 million reactions outside the training set, and is able to recover a context combination with the catalyst and at least one solvent and reagent close to the true context in the top-10 predictions in 69.6% of those cases. Qualitative evaluation on common types of reactions reveals that the model can predict the exact conditions or predict conditions that have the same functionality as the true conditions. Many failed predictions are due to highly specific reactivity or data inconsistencies. Solvent and reagent embeddings are extracted from the trained model, and the visualization of them demonstrates that these representations capture the functional similarity. The context information generated by this tool can be used to aid experimental design, improve accuracy of in silico evaluation of reactivity and pathway-level evaluation, and improve chemical synthesis processes.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscentsci.8b00357.

  • Description of computational methods, frequency vs rank plots, extended lists of reaction examples, and comparison of the model performance against baseline models (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!
  • Corresponding Author
  • Authors
    • Hanyu Gao - Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States;  Orcidhttp://orcid.org/0000-0002-6346-0739
    • Thomas J. Struble - Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States;  Orcidhttp://orcid.org/0000-0003-1695-2367
    • Connor W. Coley - Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States;  Orcidhttp://orcid.org/0000-0002-8271-8723
    • Yuran Wang - Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States
    • William H. Green - Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States;  Orcidhttp://orcid.org/0000-0003-2603-9694
  • Notes
    The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

This work was supported by the DARPA Make-It program under Contract ARO W911NF-16-2-0023. We thank Elsevier for the permission to access the Reaxys API to obtain detailed information on the reaction records. The code and models are available online at https://github.com/Coughy1991/Reaction_condition_recommendation. A user-friendly web module is available at http://askcos.mit.edu/context.

References

Click to copy section linkSection link copied!

This article references 48 other publications.

  1. 1
    Robinson, R. LXIII.–A Synthesis of Tropinone. J. Chem. Soc., Trans. 1917, 111 (0), 762– 768,  DOI: 10.1039/CT9171100762
    Google Scholar
    1
    Synthesis of tropinone
    Robinson, Robert
    Journal of the Chemical Society, Transactions (1917), 111 (), 762-8CODEN: JCHTA3; ISSN:0368-1645.
    Of the derivs. of tropinone (I) none is so suitable for the identification small amts. of (I) as dipiperonylidenetropinone (A), from (I) and piperonal boiled in alc. with aq. KOH for 15 min., yellow needles, m. 214°, sparingly sol. in most org. solvents; yield, quant., gives a blue color with H2SO4; the acetate dissolves in H2O with a bright orange-yellow color, the soln. giving cryst. ppts. with HCl, HBr, HNO3, H2SO4, (CO2H)2, and picric acid; hydrochloride (B), yellow microneedles. As a test for (I) the formation of (A) may be carried out as follows: The soln. may first be acidified and evapd. in vacuo in order to remove volatile impurities such as acetone; alc., excess of piperonal and KOH are added, and the soln. heated on the H2O bath for a few min. and poured into Et2O. This is washed with H2O and shaken with a little dil. HCl, whereupon (B) crysts. (CH2CHO)2 was prepd. according to Harries (Ber. 34, 1494(1901)) from 7 g. of the oxime in 30 cc. H2O, 50 cc. H2O added, and the HNO2 and HNO3 neutralized with pptd. CaCO3. 5 g. acetone and 4.5 g. MeNH2 in 20 cc. H2O were added and the mixt. let stand 0.5 hr., after which the soln. was tested for (I) as above and a small amt. of (B) isolated. If the mixt. was allowed to stand overnight other products were formed which interfered with the identification of (I). A more certain result was obtained when the acetone was substituted by OC(CH2CO2Et)2, letting the mixt. stand overnight, distg. off the alc., boiling with dil. H2SO4 for 0.5 hr., evapg. in a high vacuum, making alk. with KOH, and distg. off the (I) with steam and identifying as above. Best results were obtained, however, when OC(CH2CO2H)2 was used and the mixt. let stand 50 hrs., filtered, acidified with HCl, and concd. in vacuo. The residue was made alk. and distd. with steam and the distillate acidified with HCl, concd. in vacuo, made alk., and extd. with Et2O. After drying rapidly over KOH and evapg. the residue crystd. partly, the crystals being identical with a sample of (I) prepd. by the oxidation of tropine, and an additional amt. of (B) being obtained from the non-cryst. portion. Final conditions for the synthesis have not as yet been worked out, but in another expt. a yield of 42% of the theory was obtained, based on the estimation of the amt. of aldehyde used by conversion of a portion into the diphenylhydrazone and estimation of (I) by conversion into (B). It is hoped that the reaction will prove of general application, and an attempt will be made to synthesize ψ-pelletierine (II) by condensing H2C(CH2CHO)2 with MeNH2 and an acetone deriv.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaC2sXhsFShsw%253D%253D&md5=860007da88776b311b4f0070c5e56e7f
  2. 2
    Corey, E. J.; Wipke, W. T. Computer Assisted Design of Complex Organic Syntheses. Science (Washington, DC, U. S.) 1969, 166 (3902), 178– 192,  DOI: 10.1126/science.166.3902.178
    Google Scholar
    2
    Computer-assisted design of complex organic syntheses
    Corey, Elias J.; Wipke, W. Todd
    Science (Washington, DC, United States) (1969), 166 (3902), 178-92CODEN: SCIEAS; ISSN:0036-8075.
    The application of digital computers to the generation of paths for the synthesis of complex org. mols. was discussed. Given the requisite computing and graphic communications hardware a set of programs of such power as to make computerized synthetic anal. an indispensable aid is possible.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaF1MXlt1eju78%253D&md5=67a17fda766d2a14be6255cd33bfaa65
  3. 3
    Cook, A.; Johnson, A. P.; Law, J.; Mirzazadeh, M.; Ravitz, O.; Simon, A. Computer-Aided Synthesis Design: 40 Years On. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2 (1), 79– 107,  DOI: 10.1002/wcms.61
    Google Scholar
    3
    Computer-aided synthesis design. 40 years on
    Cook, Anthony; Johnson, A. Peter; Law, James; Mirzazadeh, Mahdi; Ravitz, Orr; Simon, Aniko
    Wiley Interdisciplinary Reviews: Computational Molecular Science (2012), 2 (1), 79-107CODEN: WIRCAH; ISSN:1759-0884. (Wiley-Blackwell)
    A review. The discipline of retrosynthetic anal. is now just over 40 years old. From the earliest day, attempts were made to incorporate this approach into computer programs to test the extent in which chem. perception and synthetic thinking could be formalized. Despite pioneering research efforts, computer-aided synthetic anal. failed to achieve widespread routine use by chemists, which can be attributed in part to the difficulty of building the required high-quality retrosynthetic transform databases required for credible analyses. However, with the advent over the past 25 years of large comprehensive reaction databases, work on successfully automating the construction of reliable and comprehensive reaction rule databases is promising to revitalize research in this field. This review compares and contrasts the diverse approaches taken by selected programs in both the design and implementation of mol. feature perception and reaction rule representation, and the concepts of synthetic strategy selection, representation, and execution were reviewed. In particular, the current work on automating the construction of reliable and comprehensive synthetic rule sets from available reaction databases in newer programs such as ARChem were discussed. The authors argued that the progress achieved in this aspect paves the way to a deeper exploration of computer approaches to applying strategy and control in the synthesis problem.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhvFGls7g%253D&md5=a659f7b50039d0ba6de10e4e829b6a9e
  4. 4
    Warr, W. A. A Short Review of Chemical Reaction Database Systems, Computer-Aided Synthesis Design, Reaction Prediction and Synthetic Feasibility. Mol. Inf. 2014, 33 (6–7), 469– 476,  DOI: 10.1002/minf.201400052
    Google Scholar
    4
    A Short Review of Chemical Reaction Database Systems, Computer-Aided Synthesis Design, Reaction Prediction and Synthetic Feasibility
    Warr, Wendy A.
    Molecular Informatics (2014), 33 (6-7), 469-476CODEN: MIONBS; ISSN:1868-1743. (Wiley-VCH Verlag GmbH & Co. KGaA)
    This article is the text for a pedagogical lecture to be given at the Strasbourg Summer School in Chemoinformatics in June 2104. It covers a very wide range of reaction topics including structure and reaction representation, reaction centers, atom-to-atom mapping, reaction retrieval systems, computer-aided synthesis design, retrosynthesis, reaction prediction and synthetic feasibility. In the time available, the coverage of each topic can only be cursory; the main usefulness of this article to the research community is the extensive bibliog.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXovFCmur8%253D&md5=392354d1d71f1ad31f1087d56b9afd57
  5. 5
    Engkvist, O.; Norrby, P. O.; Selmi, N.; Lam, Y. H.; Peng, Z.; Sherer, E. C.; Amberg, W.; Erhard, T.; Smyth, L. A. Computational Prediction of Chemical Reactions: Current Status and Outlook. Drug Discovery Today 2018, 23 (6), 1203– 1218,  DOI: 10.1016/j.drudis.2018.02.014
    Google Scholar
    5
    Computational prediction of chemical reactions: current status and outlook
    Engkvist, Ola; Norrby, Per-Ola; Selmi, Nidhal; Lam, Yu-hong; Peng, Zhengwei; Sherer, Edward C.; Amberg, Willi; Erhard, Thomas; Smyth, Lynette A.
    Drug Discovery Today (2018), 23 (6), 1203-1218CODEN: DDTOFS; ISSN:1359-6446. (Elsevier Ltd.)
    A review. Over the past few decades, various computational methods have become increasingly important for discovering and developing novel drugs. Computational prediction of chem. reactions is a key part of an efficient drug discovery process. In this review, we discuss important parts of this field, with a focus on utilizing reaction data to build predictive models, the existing programs for synthesis prediction, and usage of quantum mechanics and mol. mechanics (QM/MM) to explore chem. reactions. We also outline potential future developments with an emphasis on pre-competitive collaboration opportunities.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXkslSitrk%253D&md5=527a5783d106ec915e6f7afbd82db210
  6. 6
    Coley, C. W.; Green, W. H.; Jensen, K. F. Machine Learning in Computer-Aided Synthesis Planning. Acc. Chem. Res. 2018, 51 (5), 1281– 1289,  DOI: 10.1021/acs.accounts.8b00087
    Google Scholar
    6
    Machine Learning in Computer-Aided Synthesis Planning
    Coley, Connor W.; Green, William H.; Jensen, Klavs F.
    Accounts of Chemical Research (2018), 51 (5), 1281-1289CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
    Computer-aided synthesis planning (CASP) is focused on the goal of accelerating the process by which chemists decide how to synthesize small mol. compds. The ideal CASP program would take a mol. structure as input and output a sorted list of detailed reaction schemes that each connect that target to purchasable starting materials via a series of chem. feasible reaction steps. Early work in this field relied on expert-crafted reaction rules and heuristics to describe possible retrosynthetic disconnections and selectivity rules but suffered from incompleteness, infeasible suggestions, and human bias. With the relatively recent availability of large reaction corpora (such as the United States Patent and Trademark Office (USPTO), Reaxys, and SciFinder databases), consisting of millions of tabulated reaction examples, it is now possible to construct and validate purely data-driven approaches to synthesis planning. As a result, synthesis planning has been opened to machine learning techniques, and the field is advancing rapidly. In this Account, we focus on two crit. aspects of CASP and recent machine learning approaches to both challenges. First, we discuss the problem of retrosynthetic planning, which requires a recommender system to propose synthetic disconnections starting from a target mol. We describe how the search strategy, necessary to overcome the exponential growth of the search space with increasing no. of reaction steps, can be assisted through a learned synthetic complexity metric. We also describe how the recursive expansion can be performed by a straightforward nearest neighbor model that makes clever use of reaction data to generate high quality retrosynthetic disconnections. Second, we discuss the problem of anticipating the products of chem. reactions, which can be used to validate proposed reactions in a computer-generated synthesis plan (i.e., reduce false positives) to increase the likelihood of exptl. success. While we introduce this task in the context of reaction validation, its utility extends to the prediction of side products and impurities, among other applications. We describe neural network-based approaches that we and others have developed for this forward prediction task that can be trained on previously published exptl. data. Machine learning and artificial intelligence have revolutionized a no. of disciplines, not limited to image recognition, dictation, translation, content recommendation, advertising, and autonomous driving. While there is a rich history of using machine learning for structure-activity models in chem., it is only now that it is being successfully applied more broadly to org. synthesis and synthesis design. As reported in this Account, machine learning is rapidly transforming CASP, but there are several remaining challenges and opportunities, many pertaining to the availability and standardization of both data and evaluation metrics, which must be addressed by the community at large.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXosFKhsb0%253D&md5=a1ea72c55942f3c0f0a99ab080f96899
  7. 7
    Goodman, J. M. Reaction Prediction and Synthesis Design. Appl. Chemoinformatics Achiev. Futur. Oppor. 2018, 86– 105,  DOI: 10.1002/9783527806539.ch4b
    Google Scholar
    There is no corresponding record for this reference.
  8. 8
    Reaxys. https://new.reaxys.com/ (accessed on Sept 28, 2017).
    Google Scholar
    There is no corresponding record for this reference.
  9. 9
    Lowe, D. M. Patent Reaction Extractor (v1.0); 2014.
    Google Scholar
    There is no corresponding record for this reference.
  10. 10
    Szymkuć, S.; Gajewska, E. P.; Klucznik, T.; Molga, K.; Dittwald, P.; Startek, M.; Bajczyk, M.; Grzybowski, B. A. Computer-Assisted Synthetic Planning: The End of the Beginning. Angew. Chem., Int. Ed. 2016, 55 (20), 5904– 5937,  DOI: 10.1002/anie.201506101
    Google Scholar
    10
    Computer-Assisted Synthetic Planning: The End of the Beginning
    Szymkuc, Sara; Gajewska, Ewa P.; Klucznik, Tomasz; Molga, Karol; Dittwald, Piotr; Startek, Michal; Bajczyk, Michal; Grzybowski, Bartosz A.
    Angewandte Chemie, International Edition (2016), 55 (20), 5904-5937CODEN: ACIEF5; ISSN:1433-7851. (Wiley-VCH Verlag GmbH & Co. KGaA)
    Exactly half a century has passed since the launch of the first documented research project (1965 Dendral) on computer-assisted org. synthesis. Many more programs were created in the 1970s and 1980s but the enthusiasm of these pioneering days had largely dissipated by the 2000s, and the challenge of teaching the computer how to plan org. syntheses earned itself the reputation of a "mission impossible". This is quite curious given that, in the meantime, computers have "learned" many other skills that had been considered exclusive domains of human intellect and creativity-for example, machines can nowadays play chess better than human world champions and they can compose classical music pleasant to the human ear. Although there have been no similar feats in org. synthesis, this Review argues that to concede defeat would be premature. Indeed, bringing together the combination of modern computational power and algorithms from graph/network theory, chem. rules (with full stereo- and regiochem.) coded in appropriate formats, and the elements of quantum mechanics, the machine can finally be "taught" how to plan syntheses of non-trivial org. mols. in a matter of seconds to minutes. The Review begins with an overview of some basic theor. concepts essential for the big-data anal. of chem. syntheses. It progresses to the problem of optimizing pathways involving known reactions. It culminates with discussion of algorithms that allow for a completely de novo and fully automated design of syntheses leading to relatively complex targets, including those that have not been made before. Of course, there are still things to be improved, but computers are finally becoming relevant and helpful to the practice of org.-synthetic planning. Paraphrasing Churchill's famous words after the Allies' first major victory over the Axis forces in Africa, it is not the end, it is not even the beginning of the end, but it is the end of the beginning for the computer-assisted synthesis planning. The machine is here to stay.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XlvVagsbs%253D&md5=f07bb3b25f9b87c549860d7d081ed28f
  11. 11
    Segler, M. H.S.; Preuss, M.; Waller, M. P. Learning to Plan Chemical Syntheses. 2017, arXiv:1708.04202. arXiv.org e-Print archive. https://arxiv.org/abs/1708.04202.
    Google Scholar
    There is no corresponding record for this reference.
  12. 12
    Law, J.; Zsoldos, Z.; Simon, A.; Reid, D.; Liu, Y.; Khew, S. Y.; Johnson, A. P.; Major, S.; Wade, R. A.; Ando, H. Y. Route Designer: A Retrosynthetic Analysis Tool Utilizing Automated Retrosynthetic Rule Generation. J. Chem. Inf. Model. 2009, 49 (3), 593– 602,  DOI: 10.1021/ci800228y
    Google Scholar
    12
    Route Designer: A Retrosynthetic Analysis Tool Utilizing Automated Retrosynthetic Rule Generation
    Law, James; Zsoldos, Zsolt; Simon, Aniko; Reid, Darryl; Liu, Yang; Khew, Sing Yoong; Johnson, A. Peter; Major, Sarah; Wade, Robert A.; Ando, Howard Y.
    Journal of Chemical Information and Modeling (2009), 49 (3), 593-602CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
    Route Designer, version 1.0, is a new retrosynthetic anal. package that generates complete synthetic routes for target mols. starting from readily available starting materials. Rules describing retrosynthetic transformations are automatically generated from reaction databases, which ensure that the rules can be easily updated to reflect the latest reaction literature. These rules are used to carry out an exhaustive retrosynthetic anal. of the target mol., in which heuristics are used to mitigate the combinatorial explosion. Proposed routes are prioritized by an empirical rating algorithm to present a diverse profile of the most promising solns. The program runs on a server with a web-based user interface. An overview of the system is presented together with examples that illustrate Route Designer's utility.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXhsFegtL8%253D&md5=ef949298d3c3201c70bcfc2af56218ea
  13. 13
    Liu, B.; Ramsundar, B.; Kawthekar, P.; Shi, J.; Gomes, J.; Luu Nguyen, Q.; Ho, S.; Sloane, J.; Wender, P.; Pande, V. Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models. ACS Cent. Sci. 2017, 3 (10), 1103– 1113,  DOI: 10.1021/acscentsci.7b00303
    Google Scholar
    13
    Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models
    Liu, Bowen; Ramsundar, Bharath; Kawthekar, Prasad; Shi, Jade; Gomes, Joseph; Nguyen, Quang Luu; Ho, Stephen; Sloane, Jack; Wender, Paul; Pande, Vijay
    ACS Central Science (2017), 3 (10), 1103-1113CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
    We describe a fully data driven model that learns to perform a retrosynthetic reaction prediction task, which is treated as a sequence-to-sequence mapping problem. The end-to-end trained model has an encoder-decoder architecture that consists of two recurrent neural networks, which has previously shown great success in solving other sequence-to-sequence prediction tasks such as machine translation. The model is trained on 50,000 exptl. reaction examples from the United States patent literature, which span 10 broad reaction types that are commonly used by medicinal chemists. We find that our model performs comparably with a rule-based expert system baseline model, and also overcomes certain limitations assocd. with rule-based expert systems and with any machine learning approach that contains a rule-based expert system component. Our model provides an important first step toward solving the challenging problem of computational retrosynthetic anal.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXhsVahu7fI&md5=61b6213efc544c8e24e4aa8b750e28d3
  14. 14
    Bøgevig, A.; Federsel, H.-J.; Huerta, F.; Hutchings, M. G.; Kraut, H.; Langer, T.; Löw, P.; Oppawsky, C.; Rein, T.; Saller, H. Software Tool as an Idea Generator for Synthesis Prediction. Org. Process Res. Dev. 2015, 19 (2), 357– 368,  DOI: 10.1021/op500373e
    Google Scholar
    There is no corresponding record for this reference.
  15. 15
    Coley, C. W.; Rogers, L.; Green, W. H.; Jensen, K. F. Computer-Assisted Retrosynthesis Based on Molecular Similarity. ACS Cent. Sci. 2017, 3 (12), 1237– 1245,  DOI: 10.1021/acscentsci.7b00355
    Google Scholar
    15
    Computer-Assisted Retrosynthesis Based on Molecular Similarity
    Coley, Connor W.; Rogers, Luke; Green, William H.; Jensen, Klavs F.
    ACS Central Science (2017), 3 (12), 1237-1245CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
    We demonstrate mol. similarity to be a surprisingly effective metric for proposing and ranking one-step retrosynthetic disconnections based on analogy to precedent reactions. The developed approach mimics the retrosynthetic strategy defined implicitly by a corpus of known reactions without the need to encode any chem. knowledge. Using 40000 reactions from the patent literature as a knowledge base, the recorded reactants are among the top 10 proposed precursors in 74.1% of 5000 test reactions, providing strong quant. support for our methodol. Extension of the one-step strategy to multistep pathway planning is demonstrated and discussed for two exemplary drug products.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXhvVSmurfN&md5=53ca1ab17142a856afce4ffb67e6acea
  16. 16
    Segler, M. H.S.; Preuss, M.; Waller, M. P. Planning Chemical Syntheses with Deep Neural Networks and Symbolic AI. Nature 2018, 555 (7698), 604– 610,  DOI: 10.1038/nature25978
    Google Scholar
    16
    Planning chemical syntheses with deep neural networks and symbolic AI
    Segler, Marwin H. S.; Preuss, Mike; Waller, Mark P.
    Nature (London, United Kingdom) (2018), 555 (7698), 604-610CODEN: NATUAS; ISSN:0028-0836. (Nature Research)
    To plan the syntheses of small org. mols., chemists use retrosynthesis, a problem-solving technique in which target mols. are recursively transformed into increasingly simpler precursors. Computer-aided retrosynthesis would be a valuable tool but at present it is slow and provides results of unsatisfactory quality. Here, we use Monte Carlo tree search and symbolic artificial intelligence (AI) to discover retrosynthetic routes. We combined Monte Carlo tree search with an expansion policy network that guides the search, and a filter network to pre-select the most promising retrosynthetic steps. These deep neural networks were trained on essentially all reactions ever published in org. chem. Our system solves for almost twice as many mols., thirty times faster than the traditional computer-aided search method, which is based on extd. rules and hand-designed heuristics. In a double-blind AB test, chemists on av. considered our computer-generated routes to be equiv. to reported literature routes.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXmsVGqt7c%253D&md5=400e9945ff83ffe2d12278aa4c562893
  17. 17
    Kayala, M. A.; Baldi, P. F. Learning to Predict Chemical Reactions. J. Chem. Inf. Model. 2011, 51 (9), 2209– 2222,  DOI: 10.1021/ci200207y
    Google Scholar
    17
    Learning to Predict Chemical Reactions
    Kayala, Matthew A.; Azencott, Chloe-Agathe; Chen, Jonathan H.; Baldi, Pierre
    Journal of Chemical Information and Modeling (2011), 51 (9), 2209-2222CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
    Being able to predict the course of arbitrary chem. reactions is essential to the theory and applications of org. chem. Approaches to the reaction prediction problems can be organized around three poles corresponding to: (1) phys. laws; (2) rule-based expert systems; and (3) inductive machine learning. Previous approaches at these poles, resp., are not high throughput, are not generalizable or scalable, and lack sufficient data and structure to be implemented. We propose a new approach to reaction prediction utilizing elements from each pole. Using a phys. inspired conceptualization, we describe single mechanistic reactions as interactions between coarse approxns. of MOs (MOs) and use topol. and physicochem. attributes as descriptors. Using an existing rule-based system (Reaction Explorer), we derive a restricted chem. data set consisting of 1630 full multistep reactions with 2358 distinct starting materials and intermediates, assocd. with 2989 productive mechanistic steps and 6.14 million unproductive mechanistic steps. And from machine learning, we pose identifying productive mechanistic steps as a statistical ranking, information retrieval problem: given a set of reactants and a description of conditions, learn a ranking model over potential filled-to-unfilled MO interactions such that the top-ranked mechanistic steps yield the major products. The machine learning implementation follows a two-stage approach, in which we first train atom level reactivity filters to prune 94.00% of nonproductive reactions with a 0.01% error rate. Then, we train an ensemble of ranking models on pairs of interacting MOs to learn a relative productivity function over mechanistic steps in a given system. Without the use of explicit transformation patterns, the ensemble perfectly ranks the productive mechanism at the top 89.05% of the time, rising to 99.86% of the time when the top four are considered. Furthermore, the system is generalizable, making reasonable predictions over reactants and conditions which the rule-based expert does not handle. A web interface to the machine learning based mechanistic reaction predictor is accessible through our chemoinformatics portal (http://cdb.ics.uci.edu) under the Toolkits section.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhtFSks7jP&md5=1b40eaa05deaeb36da0ede14bf8d29f9
  18. 18
    Kayala, M. A.; Baldi, P. ReactionPredictor: Prediction of Complex Chemical Reactions at the Mechanistic Level Using Machine Learning. J. Chem. Inf. Model. 2012, 52 (10), 2526– 2540,  DOI: 10.1021/ci3003039
    Google Scholar
    18
    ReactionPredictor: Prediction of Complex Chemical Reactions at the Mechanistic Level Using Machine Learning
    Kayala, Matthew A.; Baldi, Pierre
    Journal of Chemical Information and Modeling (2012), 52 (10), 2526-2540CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
    Proposing reasonable mechanisms and predicting the course of chem. reactions is important to the practice of org. chem. Approaches to reaction prediction have historically used obfuscating representations and manually encoded patterns or rules. Here we present ReactionPredictor, a machine learning approach to reaction prediction that models elementary, mechanistic reactions as interactions between approx. MOs (MOs). A training data set of productive reactions known to occur at reasonable rates and yields and verified by inclusion in the literature or textbooks is derived from an existing rule-based system and expanded upon with manual curation from graduate level textbooks. Using this training data set of complex polar, hypervalent, radical, and pericyclic reactions, a two-stage machine learning prediction framework is trained and validated. In the first stage, filtering models trained at the level of individual MOs are used to reduce the space of possible reactions to consider. In the second stage, ranking models over the filtered space of possible reactions are used to order the reactions such that the productive reactions are the top ranked. The resulting model, ReactionPredictor, perfectly ranks polar reactions 78.1% of the time and recovers all productive reactions 95.7% of the time when allowing for small nos. of errors. Pericyclic and radical reactions are perfectly ranked 85.8% and 77.0% of the time, resp., rising to >93% recovery for both reaction types with a small no. of allowed errors. Decisions about which of the polar, pericyclic, or radical reaction type ranking models to use can be made with >99% accuracy. Finally, for multistep reaction pathways, we implement the first mechanistic pathway predictor using constrained tree-search to discover a set of reasonable mechanistic steps from given reactants to given products. Webserver implementations of both the single step and pathway versions of ReactionPredictor are available via the chemoinformatics portal http://cdb.ics.uci.edu/.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhtlGjtrjF&md5=3507988197eb3c3400045d0bfc939ff1
  19. 19
    Segler, M. H.S.; Waller, M. P. Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction. Chem. - Eur. J. 2017, 23 (25), 5966– 5971,  DOI: 10.1002/chem.201605499
    Google Scholar
    19
    Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction
    Segler, Marwin H. S.; Waller, Mark P.
    Chemistry - A European Journal (2017), 23 (25), 5966-5971CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
    Reaction prediction and retrosynthesis are the cornerstones of org. chem. Rule-based expert systems have been the most widespread approach to computationally solve these two related challenges to date. However, reaction rules often fail because they ignore the mol. context, which leads to reactivity conflicts. Herein, we report that deep neural networks can learn to resolve reactivity conflicts and to prioritize the most suitable transformation rules. We show that by training our model on 3.5 million reactions taken from the collective published knowledge of the entire discipline of chem., our model exhibits a top10-accuracy of 95 % in retrosynthesis and 97 % for reaction prediction on a validation set of almost 1 million reactions.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXjtlynsrw%253D&md5=e4f689a132ea6ad8713fa2d3d9422c78
  20. 20
    Jin, W.; Coley, C. W.; Barzilay, R.; Jaakkola, T. Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network , 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA; 2017; pp 2604– 2613.
    Google Scholar
    There is no corresponding record for this reference.
  21. 21
    Coley, C. W.; Barzilay, R.; Jaakkola, T. S.; Green, W. H.; Jensen, K. F. Prediction of Organic Reaction Outcomes Using Machine Learning. ACS Cent. Sci. 2017, 3 (5), 434– 443,  DOI: 10.1021/acscentsci.7b00064
    Google Scholar
    21
    Prediction of Organic Reaction Outcomes Using Machine Learning
    Coley, Connor W.; Barzilay, Regina; Jaakkola, Tommi S.; Green, William H.; Jensen, Klavs F.
    ACS Central Science (2017), 3 (5), 434-443CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
    Computer assistance in synthesis design has existed for over 40 years, yet retrosynthesis planning software has struggled to achieve widespread adoption. One crit. challenge in developing high-quality pathway suggestions is that proposed reaction steps often fail when attempted in the lab., despite initially seeming viable. The true measure of success for any synthesis program is whether the predicted outcome matches what is obsd. exptl. We report a model framework for anticipating reaction outcomes that combines the traditional use of reaction templates with the flexibility in pattern recognition afforded by neural networks. Using 15 000 exptl. reaction records from granted United States patents, a model is trained to select the major (recorded) product by ranking a self-generated list of candidates where one candidate is known to be the major product. Candidate reactions are represented using a unique edit-based representation that emphasizes the fundamental transformation from reactants to products, rather than the constituent mols.' overall structures. In a 5-fold cross-validation, the trained model assigns the major product rank 1 in 71.8% of cases, rank ≤3 in 86.7% of cases, and rank ≤5 in 90.8% of cases.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXmtVyqtb0%253D&md5=cc3e8a50d8dd9d68e294a21a558200dc
  22. 22
    Schwaller, P.; Laino, T. “Found in Translation”: Predicting Outcomes of Complex Organic Chemistry Reactions Using Neural Sequence-to-Sequence Models. 2017, arXiv:1711.04810. arXiv.org e-Print archive. https://arxiv.org/abs/1711.04810.
    Google Scholar
    There is no corresponding record for this reference.
  23. 23
    Reizman, B. J.; Jensen, K. F. Simultaneous Solvent Screening and Reaction Optimization in Microliter Slugs. Chem. Commun. 2015, 51 (68), 13290– 13293,  DOI: 10.1039/C5CC03651H
    Google Scholar
    23
    Simultaneous solvent screening and reaction optimization in microliter slugs
    Reizman, Brandon J.; Jensen, Klavs F.
    Chemical Communications (Cambridge, United Kingdom) (2015), 51 (68), 13290-13293CODEN: CHCOFS; ISSN:1359-7345. (Royal Society of Chemistry)
    An automated, continuous flow droplet screening system is presented, enabling real-time simultaneous solvent and continuous variable optimization. An optimal design of expts. strategy is applied to the alkylation of 1,2-diaminocyclohexane in 16 μL droplets, with scale-up demonstrated. Anal. of segmented flow results suggests correlation of yield with solvent hydrogen bond basicity.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXht1SqtL7O&md5=a27eeadd9328ade293e218e0167c87b8
  24. 24
    Sans, V.; Porwol, L.; Dragone, V.; Cronin, L. A Self Optimizing Synthetic Organic Reactor System Using Real-Time In-Line NMR Spectroscopy. Chem. Sci. 2015, 6 (2), 1258– 1264,  DOI: 10.1039/C4SC03075C
    Google Scholar
    24
    A self optimizing synthetic organic reactor system using real-time in-line NMR spectroscopy
    Sans, Victor; Porwol, Luzian; Dragone, Vincenza; Cronin, Leroy
    Chemical Science (2015), 6 (2), 1258-1264CODEN: CSHCCN; ISSN:2041-6520. (Royal Society of Chemistry)
    A configurable platform for synthetic chem. incorporating an in-line bench-top NMR capable of monitoring and controlling org. reactions in real-time is discussed. The platform is controlled by a modular LabView software control system for hardware, NMR, data anal., and feedback optimization. Using this platform, real-time advanced structural characterization of reaction mixts., including 19F, 13C, DEPT, 2-dimensional NMR spectroscopy (COSY, HSQC, 19F-COSY), are reported for the first time. The potential of this technique was demonstrated by optimizing a catalytic org. reaction in real-time, showing its applicability to self-optimizing systems using criteria such as stereo-selectivity, multi-nuclear measurements, or 2-dimensional correlations.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXhvFeisb%252FO&md5=e3d6d3dbfb1ba77055defb4078800dc1
  25. 25
    Holmes, N.; Akien, G. R.; Savage, R. J.D.; Stanetty, C.; Baxendale, I. R.; Blacker, A. J.; Taylor, B. A.; Woodward, R. L.; Meadows, R. E.; Bourne, R. A. Reaction Chemistry & Engineering Online Quantitative Mass Spectrometry for the Reactors †. React. Chem. Eng. 2016, 1 (1), 96– 100,  DOI: 10.1039/C5RE00083A
    Google Scholar
    25
    Online quantitative mass spectrometry for the rapid adaptive optimisation of automated flow reactors
    Holmes, Nicholas; Akien, Geoffrey R.; Savage, Robert J. D.; Stanetty, Christian; Baxendale, Ian R.; Blacker, A. John; Taylor, Brian A.; Woodward, Robert L.; Meadows, Rebecca E.; Bourne, Richard A.
    Reaction Chemistry & Engineering (2016), 1 (1), 96-100CODEN: RCEEBW; ISSN:2058-9883. (Royal Society of Chemistry)
    An automated continuous reactor for the synthesis of org. compds., which uses online mass spectrometry (MS) for reaction monitoring and product quantification, is presented. Quant. and rapid MS monitoring was developed and calibrated using HPLC. The amidation of Me nicotinate with aq. MeNH2 was optimized using design of expts. and a self-optimization algorithm approach to produce >93% yield.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhtlSlt7zM&md5=a8755576a3a4e7c07c10d7f1f3f675f2
  26. 26
    Reizman, B. J.; Jensen, K. F. Feedback in Flow for Accelerated Reaction Development. Acc. Chem. Res. 2016, 49 (9), 1786– 1796,  DOI: 10.1021/acs.accounts.6b00261
    Google Scholar
    26
    Feedback in Flow for Accelerated Reaction Development
    Reizman, Brandon J.; Jensen, Klavs F.
    Accounts of Chemical Research (2016), 49 (9), 1786-1796CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
    The pharmaceutical industry is investing in continuous flow and high-throughput experimentation as tools for rapid process development accelerated scale-up. Coupled with automation, these technologies offer the potential for comprehensive reaction characterization and optimization, but with the cost of conducting exhaustive multifactor screens. Automated feedback in flow offers researchers an alternative strategy for efficient characterization of reactions based on the use of continuous technol. to control chem. reaction conditions and optimize in lieu of screening. Optimization with feedback allows expts. to be conducted where the most information can be gained from the chem., enabling product yields to be maximized and kinetic models to be generated while the total no. of expts. is minimized. This account opens by reviewing select examples of feedback optimization in flow and applications to chem. research. Systems in the literature are classified into (i) deterministic black box optimization systems that do not model the reaction system and are therefore limited in the utility of results for scale-up, (ii) deterministic model-based optimization systems from which reaction kinetics and/or mechanisms can be automatically evaluated, and (iii) stochastic systems. Though diverse in application, flow feedback systems have predominantly focused upon the optimization of continuous variables, i.e., variables such as time, temp., and concn. that can be ramped from one expt. to the next. Unfortunately, this implies that the screening of discrete variables such as catalyst, ligand, or solvent generally does not factor into automated flow optimization, resulting in incomplete process knowledge. Herein, a system and strategy is presented developed for optimizing discrete and continuous variables of a chem. reaction simultaneously. The approach couples automated feedback with high-throughput reaction screening in droplet flow microfluidics. This account details the system configuration for on-demand creation of sub-20 μL droplets with interchangeable reagents and catalysts. These droplets are reacted in a fully automated microfluidic system and analyzed online by LC/MS. Feeding back from the online anal. results, a design of expts. (DoE)-based adaptive response surface algorithm is employed that deductively removes candidate reagents from the optimization as optimal reaction conditions are refined, leading to rapid convergence. Using the automated optimization platform, case studies are presented for solvent selection in a competitive alkylation chem. and for catalyst-ligand selection in heteroarom. Suzuki-Miyaura cross-coupling chemistries. For the monoalkylation of trans-1,2-diaminocyclohexane, polar aprotic solvents at moderate temps. are shown to be favorable, with optimality accurately identified with DMSO as the solvent in 67 expts. For Suzuki-Miyaura cross-couplings, the optimality of precatalysts and continuous variable conditions are obsd. to change in accordance with the coupling reagents, providing insights into catalyst behavior in the context of the reaction mechanism. Future opportunities in automated reaction development include the incorporation of chemoinformatics for faster anal. and machine-learning algorithms to guide and optimize the synthesis. Adoption of this technol. stands to reduce graduate student and postdoc time on routine tasks in the lab., while feeding back knowledge used to guide new research directions. Moreover, the application of this technol. in industry promises to lessen the cost and time assocd. with advancing pharmaceutical mols. through development and scale-up.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28Xhtlaru7%252FI&md5=ddaef4153972b8b045df12015e5b5d14
  27. 27
    Baumgartner, L. M.; Coley, C. W.; Reizman, B. J.; Gao, K. W.; Jensen, K. F. Optimum Catalyst Selection over Continuous and Discrete Process Variables with a Single Droplet Microfluidic Reaction Platform. React. Chem. Eng. 2018, 3 (3), 301– 311,  DOI: 10.1039/C8RE00032H
    Google Scholar
    27
    Optimum catalyst selection over continuous and discrete process variables with a single droplet microfluidic reaction platform
    Baumgartner, Lorenz M.; Coley, Connor W.; Reizman, Brandon J.; Gao, Kevin W.; Jensen, Klavs F.
    Reaction Chemistry & Engineering (2018), 3 (3), 301-311CODEN: RCEEBW; ISSN:2058-9883. (Royal Society of Chemistry)
    A mixed-integer nonlinear program (MINLP) algorithm to optimize catalyst turnover no. (TON) and product yield by simultaneously modulating discrete variables-catalyst types-and continuous variables-temp., residence time, and catalyst loading-was implemented and validated. Several simulated case studies, with and without random measurement error, demonstrate the algorithm's robustness in finding optimal conditions in the presence of side reactions and other complicating nonlinearities. This algorithm was applied to the real-time optimization of a Suzuki-Miyaura cross-coupling reaction in an automated microfluidic reaction platform comprising a liq. handler, an oscillatory flow reactor, and an online LC/MS. The algorithm, based on a combination of branch and bound and adaptive response surface methods, identified exptl. conditions that maximize TON subject to a yield constraint from a pool of eight catalyst candidates in just 60 expts., considerably fewer than a previous version of the algorithm.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXntlCntbo%253D&md5=a77e6dec5646d61ecc5eb42d04786fb4
  28. 28
    Kamlet, M. J.; Abboud, J. L.M.; Abraham, M. H.; Taft, R. W. Linear Solvation Energy Relationships. 23. A Comprehensive Collection of the Solvatochromic Parameters,.Pi.*,.Alpha., and.Beta., and Some Methods for Simplifying the Generalized Solvatochromic Equation. J. Org. Chem. 1983, 48 (17), 2877– 2887,  DOI: 10.1021/jo00165a018
    Google Scholar
    28
    Linear solvation energy relationships. 23. A comprehensive collection of the solvatochromic parameters, π*, α, and β, and some methods for simplifying the generalized solvatochromic equation
    Kamlet, Mortimer J.; Abboud, Jose Luis M.; Abraham, Michael H.; Taft, R. W.
    Journal of Organic Chemistry (1983), 48 (17), 2877-87CODEN: JOCEAH; ISSN:0022-3263.
    A generalized equation for linear solvation energy relations or complexation energy relations is developed which involves 6 terms: π* (a solvent dipolarity-polarizability term), α (solvent hydrogen-bond acceptor term), β (solvent hydrogen-bond donor term), δ (solvent polarizability correction term), δH (Hildebrand soly. parameter), and ξ. This equation is reduced to a more manageable form by a judicious choice of solvents and reactants or indicators. One-, two- or three-parameter LFER involving different combinations of the above parameters and various types of physicochem. properties are obsd. A comprehensive collection of π*, α, and β for 217 solvents is presented.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaL3sXkvVOgsbc%253D&md5=99b60c3cc817d5508b28d85f3ad6c8a1
  29. 29
    Struebing, H.; Ganase, Z.; Karamertzanis, P. G.; Siougkrou, E.; Haycock, P.; Piccione, P. M.; Armstrong, A.; Galindo, A.; Adjiman, C. S. Computer-Aided Molecular Design of Solvents for Accelerated Reaction Kinetics. Nat. Chem. 2013, 5 (11), 952– 957,  DOI: 10.1038/nchem.1755
    Google Scholar
    29
    Computer-aided molecular design of solvents for accelerated reaction kinetics
    Struebing, Heiko; Ganase, Zara; Karamertzanis, Panagiotis G.; Siougkrou, Eirini; Haycock, Peter; Piccione, Patrick M.; Armstrong, Alan; Galindo, Amparo; Adjiman, Claire S.
    Nature Chemistry (2013), 5 (11), 952-957CODEN: NCAHBB; ISSN:1755-4330. (Nature Publishing Group)
    Solvents can significantly alter the rates and selectivity of liq.-phase org. reactions, often hindering the development of new synthetic routes or, if chosen wisely, facilitating routes by improving rates and selectivities. To address this challenge, a systematic methodol. is proposed that quickly identifies improved reaction solvents by combining quantum mech. computations of the reaction rate const. in a few solvents with a computer-aided mol. design (CAMD) procedure. The approach allows the identification of a high-performance solvent within a very large set of possible mols. The validity of the authors' CAMD approach is demonstrated through application to a classical nucleophilic substitution reaction for the study of solvent effects, the Menschutkin reaction. The results were validated successfully by in situ kinetic expts. A space of 1,341 solvents was explored in silico, but required quantum-mech. calcns. of the rate const. in only nine solvents, and uncovered a solvent that increases the rate const. by 40%.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3sXhsV2jsrbP&md5=2fcd40cb79d2e6db80227791a628da0b
  30. 30
    Marcou, G.; Aires De Sousa, J.; Latino, D. A.R.S.; De Luca, A.; Horvath, D.; Rietsch, V.; Varnek, A. Expert System for Predicting Reaction Conditions: The Michael Reaction Case. J. Chem. Inf. Model. 2015, 55 (2), 239– 250,  DOI: 10.1021/ci500698a
    Google Scholar
    30
    Expert System for Predicting Reaction Conditions: The Michael Reaction Case
    Marcou, G.; Aires de Sousa, J.; Latino, D. A. R. S.; de Luca, A.; Horvath, D.; Rietsch, V.; Varnek, A.
    Journal of Chemical Information and Modeling (2015), 55 (2), 239-250CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
    A generic chem. transformation may often be achieved under various synthetic conditions. However, for any specific reagents, only one or a few among the reported synthetic protocols may be successful. For example, Michael β-addn. reactions may proceed under different choices of solvent (e.g., hydrophobic, aprotic polar, protic) and catalyst (e.g., Bronsted acid, Lewis acid, Lewis base, etc.). Chemoinformatics methods could be efficiently used to establish a relationship between the reagent structures and the required reaction conditions, which would allow synthetic chemists to waste less time and resources in trying out various protocols in search for the appropriate one. In order to address this problem, a no. of 2-classes classification models have been built on a set of 198 Michael reactions retrieved from literature. Trained models discriminate between processes that are compatible and resp. processes not feasible under a specific reaction condition option (feasible or not with a Lewis acid catalyst, feasible or not in hydrophobic solvent, etc.). Eight distinct models were built to decide the compatibility of a Michael addn. process with each considered reaction condition option, while a ninth model was aimed to predict whether the assumed Michael addn. is feasible at all. Different machine-learning methods (Support Vector Machine, Naive Bayes, and Random Forest) in combination with different types of descriptors (ISIDA fragments issued from Condensed Graphs of Reactions, MOLMAP, Electronic Effect Descriptors, and Chem. Development Kit computed descriptors) have been used. Models have good predictive performance in 3-fold cross-validation done three times: balanced accuracy varies from 0.7 to 1. Developed models are available for the users at http://infochim.u-strasbg.fr/webserv/VSEngine.html. Eventually, these were challenged to predict feasibility conditions for ∼50 novel Michael reactions from the eNovalys database (originally from patent literature).
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXnvFGnsQ%253D%253D&md5=205b5b0bd2af7f06f63174e2f6f799b1
  31. 31
    Lin, A. I.; Madzhidov, T. I.; Klimchuk, O.; Nugmanov, R. I.; Antipin, I. S.; Varnek, A. Automatized Assessment of Protective Group Reactivity: A Step toward Big Reaction Data Analysis. J. Chem. Inf. Model. 2016, 56 (11), 2140– 2148,  DOI: 10.1021/acs.jcim.6b00319
    Google Scholar
    31
    Automatized Assessment of Protective Group Reactivity: A Step Toward Big Reaction Data Analysis
    Lin, Arkadii I.; Madzhidov, Timur I.; Klimchuk, Olga; Nugmanov, Ramil I.; Antipin, Igor S.; Varnek, Alexandre
    Journal of Chemical Information and Modeling (2016), 56 (11), 2140-2148CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
    The authors report a new method to assess protective groups (PGs) reactivity as a function of reaction conditions (catalyst, solvent) using raw reaction data. It is based on an intuitive similarity principle for chem. reactions: similar reactions proceed under similar conditions. Tech., reaction similarity can be assessed using the Condensed Graph of Reaction (CGR) approach representing an ensemble of reactants and products as a single mol. graph, i.e., as a pseudomol. for which mol. descriptors or fingerprints can be calcd. CGR-based inhouse tools were used to process data for 142,111 catalytic hydrogenation reactions extd. from the Reaxys database. Results reveal some contradictions with famous Greene's Reactivity Charts based on manual expert anal. Models developed in this study show high accuracy (∼90%) for predicting optimal exptl. conditions of protective group deprotection.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhslemsL%252FM&md5=734545a7f2a3b34f49af6713d443b6be
  32. 32
    Segler, M. H.S.; Waller, M. P. Modelling Chemical Reasoning to Predict and Invent Reactions. Chem. - Eur. J. 2017, 23 (25), 6118– 6128,  DOI: 10.1002/chem.201604556
    Google Scholar
    32
    Modelling Chemical Reasoning to Predict and Invent Reactions
    Segler, Marwin H. S.; Waller, Mark P.
    Chemistry - A European Journal (2017), 23 (25), 6118-6128CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
    The ability to reason beyond established knowledge allows org. chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chem. reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph contg. 14.4 million mols. and 8.2 million binary reactions, which represents the bulk of all chem. reactions ever published in the scientific literature. Our model outperforms a rule-based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data-driven model generalises even beyond known reaction types, and is thus capable of effectively (re-)discovering novel transformations (even including transition metal-catalyzed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub-second time frame, the model can be used as a high-throughput generator of reaction hypotheses for reaction discovery.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXislersw%253D%253D&md5=ac3b304ec62b8d90110b7722305e2b3d
  33. 33
    Mikie Kanada, R.; Taniguchi, T.; Ogasawara, K. Asymmetric Hydrogen Transfer Protocol for a Synthesis of (+)-Frontalin and (−)-Malyngolide. Tetrahedron Lett. 2000, 41 (19), 3631– 3635,  DOI: 10.1016/S0040-4039(00)00430-5
    Google Scholar
    There is no corresponding record for this reference.
  34. 34
    Faroux-Corlay, B.; Clary, L.; Gadras, C.; Hammache, D.; Greiner, J.; Santaella, C.; Aubertin, A. M.; Vierling, P.; Fantini, J. Synthesis of Single- and Double-Chain Fluorocarbon and Hydrocarbon Galactosyl Amphiphiles and Their Anti-HIV-1 Activity. Carbohydr. Res. 2000, 327 (3), 223– 260,  DOI: 10.1016/S0008-6215(00)00055-0
    Google Scholar
    There is no corresponding record for this reference.
  35. 35
    Wang, H.; Yu, S. Synthesis of Isoquinolones Using Visible-Light-Promoted Denitrogenative Alkyne Insertion of 1,2,3-Benzotriazinones. Org. Lett. 2015, 17 (17), 4272– 4275,  DOI: 10.1021/acs.orglett.5b01960
    Google Scholar
    35
    Synthesis of Isoquinolones Using Visible-Light-Promoted Denitrogenative Alkyne Insertion of 1,2,3-Benzotriazinones
    Wang, Hao; Yu, Shouyun
    Organic Letters (2015), 17 (17), 4272-4275CODEN: ORLEF7; ISSN:1523-7052. (American Chemical Society)
    A visible-light-promoted regioselective denitrogenative insertion of terminal alkynes into 1,2,3-benzotriazinones is reported. This mechanistically novel process allows the synthesis of substituted isoquinolones in satisfactory isolated yields (24 examples, 46-84% yield) at room temp. under visible-light irradn. with the assistance of a photocatalyst. The proposed single-electron-transfer pathway was supported by TEMPO trapping, radical clock expts., and Stern-Volmer anal.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhtlOjur3K&md5=86034a6db6d0456dbe780e5faec49e63
  36. 36
    Li, K.; Zeng, Y.; Neuenswander, B. Sequential Pd (II) -Pd (0) Catalysis for the Rapid Synthesis of Coumarins. J. Org. Chem. 2005, 70 (16), 6515– 6518,  DOI: 10.1021/jo050671l
    Google Scholar
    There is no corresponding record for this reference.
  37. 37
    Mavunkel, B.; Xu, Y.; Goyal, B.; Lim, D.; Lu, Q.; Chen, Z.; Wang, D.-X.; Higaki, J.; Chakraborty, I.; Liclican, A. Pyrimidine-Based Inhibitors of CaMKIIδ. Bioorg. Med. Chem. Lett. 2008, 18 (7), 2404– 2408,  DOI: 10.1016/j.bmcl.2008.02.056
    Google Scholar
    There is no corresponding record for this reference.
  38. 38
    Lautens, M.; Maddess, M. L. Chemoselective Cross Metathesis of Bishomoallylic Alcohols : Rapid Access to Fragment A of the Cryptophycins. Supplementary Material The Following Includes Representative Experimental Procedures and Details for Isolation of Compounds. Full Characterisat. Org. Lett. 2004, 6 (12), 1883– 1886,  DOI: 10.1021/ol049883f
    Google Scholar
    There is no corresponding record for this reference.
  39. 39
    Krüger, T.; Vorndran, K.; Linker, T. Regioselective Arene Functionalization: Simple Substitution of Carboxylate by Alkyl Groups. Chem. - Eur. J. 2009, 15 (44), 12082– 12091,  DOI: 10.1002/chem.200901774
    Google Scholar
    39
    Regioselective Arene Functionalization: Simple Substitution of Carboxylate by Alkyl Groups
    Krueger, Tobias; Vorndran, Katja; Linker, Torsten
    Chemistry - A European Journal (2009), 15 (44), 12082-12091CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
    Arenes with various alkyl side-chains were synthesized in high yields and excellent regioselectivities. The carboxylate group in toluic and naphthoic acids was conveniently substituted by alkyl halides by Birch redn. and subsequent decarbonylation. The method is characterized by inexpensive starting materials and reagents, and methylation of arenes was realized. Besides simple alkyl substituents, the scope of arene functionalization was extended by benzyl as well as fluoro-, amino-, and ester-contg. alkyl groups. The alkylation of 1-naphthoic acid during Birch redn. can be controlled by the addn. of tert-butanol. This allowed the regioselective synthesis of mono and bis-substituted naphthalenes from the same starting material.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXhtlyqu77F&md5=b143b98bc1a436b16bf909a9347629dc
  40. 40
    Palmes, J. A.; Paioti, P. H.S.; De Souza, L. P.; Aponick, A. PdII-Catalyzed Spiroketalization of Ketoallylic Diols. Chem. - Eur. J. 2013, 19 (35), 11613– 11621,  DOI: 10.1002/chem.201301723
    Google Scholar
    40
    PdII-Catalyzed Spiroketalization of Ketoallylic Diols
    Palmes, Jean A.; Paioti, Paulo H. S.; de Souza, Leonardo Perez; Aponick, Aaron
    Chemistry - A European Journal (2013), 19 (35), 11613-11621CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
    A high-yielding stereoselective method for forming spiroketals from simple ketone-allylic diol derivs. was reported. Using catalytic [PdCl2(MeCN)2] in THF at 0° these dehydration-cyclization reactions require only mild conditions to produce vinyl-substituted spiroketals in high yields after brief reaction times with water as the only byproduct. Using this method, the stereochem. information embedded at the nucleophile is transmitted down-the-chain and efficiently sets the stereochem. at both the anomeric carbon atom and the newly formed allylic stereocenter. The title compds. thus formed included a spiroketal (I) and related substances, such as ro[5.5]undecane derivs. and 1,6-dioxaspiro[4.5]decane derivs., carbohydrate monosaccharide analogs, such as a 5,9-anhydro-2,3,4-trideoxy-D-gluco-5-deculo-5,1-pyranose deriv. The synthesis of the target compds. was achieved by a palladium-catalyzed cyclization of a chiral 3,13-dihydroxy-11-tetradecen-7-one deriv. (II) and similar compds., for example (8E)-1,10-dihydroxy-8-decen-5-one.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3sXhtFShtLrN&md5=4271b9729abcbf0817cd65ea2cd5fb49
  41. 41
    Liu, J.; Fitzgerald, A. E.; Mani, N. S. Facile Assembly of Fused Benzo[4,5]Furo Heterocycles. J. Org. Chem. 2008, 73 (7), 2951– 2954,  DOI: 10.1021/jo8000595
    Google Scholar
    There is no corresponding record for this reference.
  42. 42
    Schaub, C.; Müller, B.; Schmidt, R. R. Sialyltransferase Inhibitors Based on CMP-Quinic Acid. Eur. J. Org. Chem. 2000, 2000 (9), 1745– 1758,  DOI: 10.1002/(SICI)1099-0690(200005)2000:9<1745::AID-EJOC1745>3.0.CO;2-8
    Google Scholar
    There is no corresponding record for this reference.
  43. 43
    Mikolov, T.; Sutskever, I.; Chen, K.; Corrado, G. S.; Dean, J. Distributed Representations of Words and Phrases and Their Compositionality. In Advances in neural information processing systems; NIPS: Lake Tahoe, 2013; pp 3111– 3119.
    Google Scholar
    There is no corresponding record for this reference.
  44. 44
    García-Alonso, C. R.; Pérez-Naranjo, L. M.; Fernández-Caballero, J. C. Multiobjective evolutionary algorithms to identify highly autocorrelated areas: the case of spatial distribution in financially compromised farms. Ann. Oper. Res. 2014, 219 (1), 187– 202,  DOI: 10.1007/s10479-011-0841-3
    Google Scholar
    There is no corresponding record for this reference.
  45. 45
    Open-source. RDKit: Open-Source Cheminformatics Software , 2006 (accessed on Sept 28, 2017).
    Google Scholar
    There is no corresponding record for this reference.
  46. 46
    Schneider, N.; Lowe, D. M.; Sayle, R. A.; Landrum, G. A. Development of a Novel Fingerprint for Chemical Reactions and Its Application to Large-Scale Reaction Classification and Similarity. J. Chem. Inf. Model. 2015, 55 (1), 39– 53,  DOI: 10.1021/ci5006614
    Google Scholar
    46
    Development of a Novel Fingerprint for Chemical Reactions and Its Application to Large-Scale Reaction Classification and Similarity
    Schneider, Nadine; Lowe, Daniel M.; Sayle, Roger A.; Landrum, Gregory A.
    Journal of Chemical Information and Modeling (2015), 55 (1), 39-53CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
    Fingerprint methods applied to mols. have proven to be useful for similarity detn. and as inputs to machine-learning models. Here, we present the development of a new fingerprint for chem. reactions and validate its usefulness in building machine-learning models and in similarity assessment. Our final fingerprint is constructed as the difference of the atom-pair fingerprints of products and reactants and includes agents via calcd. physicochem. properties. We validated the fingerprints on a large data set of reactions text-mined from granted United States patents from the last 40 years that have been classified using a substructure-based expert system. We applied machine learning to build a 50-class predictive model for reaction-type classification that correctly predicts 97% of the reactions in an external test set. Impressive accuracies were also obsd. when applying the classifier to reactions from an inhouse electronic lab. notebook. The performance of the novel fingerprint for assessing reaction similarity was evaluated by a cluster anal. that recovered 48 out of 50 of the reaction classes with a median F-score of 0.63 for the clusters. The data sets used for training and primary validation as well as all python scripts required to reproduce the anal. are provided in the Supporting Information.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXitFGrsrrE&md5=694e4122d2642de22065e46d7a0826cc
  47. 47
    Williams, R. J.; Zipser, D. A Learning Algorithm for Continually Running Fully Recurrent Neural Networks. Neural Comput. 1989, 1 (2), 270– 280,  DOI: 10.1162/neco.1989.1.2.270
    Google Scholar
    There is no corresponding record for this reference.
  48. 48
    Gobbi, A.; Poppinger, D. Genetic Optimization of Combinatorial Libraries. Biotechnol. Bioeng. 1998, 61 (1), 47– 54,  DOI: 10.1002/(SICI)1097-0290(199824)61:1<47::AID-BIT9>3.0.CO;2-Z
    Google Scholar
    48
    Genetic optimization of combinatorial libraries
    Gobbi, Alberto; Poppinger, Dieter
    Biotechnology and Bioengineering (1998), 61 (1), 47-53CODEN: BIBIAU; ISSN:0006-3592. (John Wiley & Sons, Inc.)
    Most agrochem. and pharmaceutical companies have set up high-throughput screening programs which require large nos. of compds. to screen. Combinatorial libraries provide an attractive way to deliver these compds. A single combinatorial library with four variable positions can yield >1012 potential compds., if one assumes that ∼1000 reagents are available for each position. This is far more than any high-throughput screening facility can afford to screen. We have proposed a method for iterative compd. selection from large databases, which identifies the most active compds. by examg. only a small fraction of the database. In this article, we describe the extension of this method to the problem of selecting compds. from large combinatorial libraries.
    >> More from SciFinder ®
    https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK1cXisVeksb8%253D&md5=3cfe243bf7cea6951886c5551f5ab3c6

Cited By

Click to copy section linkSection link copied!
Citation Statements
Explore this article's citation statements on scite.ai
powered by  Scite

This article is cited by 289 publications.

  1. Matthew A. McDonald, Brent A. Koscher, Richard B. Canty, Jason Zhang, Angelina Ning, Klavs F. Jensen. Bayesian Optimization over Multiple Experimental Fidelities Accelerates Automated Discovery of Drug Molecules. ACS Central Science 2025, 11 (2) , 346-356. https://doi.org/10.1021/acscentsci.4c01991
  2. Lanxin Long, Rui Li, Jian Zhang. Artificial Intelligence in Retrosynthesis Prediction and its Applications in Medicinal Chemistry. Journal of Medicinal Chemistry 2025, 68 (3) , 2333-2355. https://doi.org/10.1021/acs.jmedchem.4c02749
  3. Dong-Zhi Li, Xue-Qing Gong. Challenges with Literature-Derived Data in Machine Learning for Yield Prediction: A Case Study on Pd-Catalyzed Carbonylation Reactions. The Journal of Physical Chemistry A 2024, 128 (48) , 10423-10430. https://doi.org/10.1021/acs.jpca.4c05489
  4. Alexander S. Shved, Blake E. Ocampo, Elena S. Burlova, Casey L. Olen, N. Ian Rinehart, Scott E. Denmark. molli: A General Purpose Python Toolkit for Combinatorial Small Molecule Library Generation, Manipulation, and Feature Extraction. Journal of Chemical Information and Modeling 2024, 64 (21) , 8083-8090. https://doi.org/10.1021/acs.jcim.4c00424
  5. Ajnabiul Hoque, Mihir Surve, Shivaram Kalyanakrishnan, Raghavan B. Sunoj. Reinforcement Learning for Improving Chemical Reaction Performance. Journal of the American Chemical Society 2024, 146 (41) , 28250-28267. https://doi.org/10.1021/jacs.4c08866
  6. Vincent Fan, Yujie Qian, Alex Wang, Amber Wang, Connor W. Coley, Regina Barzilay. OpenChemIE: An Information Extraction Toolkit for Chemistry Literature. Journal of Chemical Information and Modeling 2024, 64 (14) , 5521-5534. https://doi.org/10.1021/acs.jcim.4c00572
  7. Samuel T. Cahill, Joseph E. B. Young, Max Howe, Ryan Clark, Andrew F. Worrall, Malcolm I. Stewart. Assignment of Regioisomers Using Infrared Spectroscopy: A Python Coding Exercise in Data Processing and Machine Learning. Journal of Chemical Education 2024, 101 (7) , 2925-2932. https://doi.org/10.1021/acs.jchemed.4c00295
  8. Liang Gao, Jiaping Lin, Liquan Wang, Lei Du. Machine Learning-Assisted Design of Advanced Polymeric Materials. Accounts of Materials Research 2024, 5 (5) , 571-584. https://doi.org/10.1021/accountsmr.3c00288
  9. Eisuke Sato, Gaku Tachiwaki, Mayu Fujii, Koichi Mitsudo, Takashi Washio, Shinobu Takizawa, Seiji Suga. Electrochemical Carbon-Ferrier Rearrangement Using a Microflow Reactor and Machine Learning-Assisted Exploration of Suitable Conditions. Organic Process Research & Development 2024, 28 (5) , 1422-1429. https://doi.org/10.1021/acs.oprd.2c00267
  10. Daniel S. Wigh, Joe Arrowsmith, Alexander Pomberger, Kobi C. Felton, Alexei A. Lapkin. ORDerly: Data Sets and Benchmarks for Chemical Reaction Data. Journal of Chemical Information and Modeling 2024, 64 (9) , 3790-3798. https://doi.org/10.1021/acs.jcim.4c00292
  11. Yuheng Ding, Bo Qiang, Qixuan Chen, Yiqiao Liu, Liangren Zhang, Zhenming Liu. Exploring Chemical Reaction Space with Machine Learning Models: Representation and Feature Perspective. Journal of Chemical Information and Modeling 2024, 64 (8) , 2955-2970. https://doi.org/10.1021/acs.jcim.4c00004
  12. Hiroaki Okada, Satoshi Maeda. On Accelerating Substrate Optimization Using Computational Gibbs Energy Barriers: A Numerical Consideration Utilizing a Computational Data Set. ACS Omega 2024, 9 (6) , 7123-7131. https://doi.org/10.1021/acsomega.3c09066
  13. Varvara Voinarovska, Mikhail Kabeshov, Dmytro Dudenko, Samuel Genheden, Igor V. Tetko. When Yield Prediction Does Not Yield Prediction: An Overview of the Current Challenges. Journal of Chemical Information and Modeling 2024, 64 (1) , 42-56. https://doi.org/10.1021/acs.jcim.3c01524
  14. Erin C. Day, Supraja S. Chittari, Matthew P. Bogen, Abigail S. Knight. Navigating the Expansive Landscapes of Soft Materials: A User Guide for High-Throughput Workflows. ACS Polymers Au 2023, 3 (6) , 406-427. https://doi.org/10.1021/acspolymersau.3c00025
  15. Alessandra Toniato, Alain C. Vaucher, Marzena Maria Lehmann, Torsten Luksch, Philippe Schwaller, Marco Stenta, Teodoro Laino. Fast Customization of Chemical Language Models to Out-of-Distribution Data Sets. Chemistry of Materials 2023, 35 (21) , 8806-8815. https://doi.org/10.1021/acs.chemmater.3c01406
  16. Hossam Nada, Anam Rana Gul, Ahmed Elkamhawy, Sungdo Kim, Minkyoung Kim, Yongseok Choi, Tae Jung Park, Kyeong Lee. Machine Learning-Based Approach to Developing Potent EGFR Inhibitors for Breast Cancer─Design, Synthesis, and In Vitro Evaluation. ACS Omega 2023, 8 (35) , 31784-31800. https://doi.org/10.1021/acsomega.3c02799
  17. William Finnigan, Max Lubberink, Lorna J. Hepworth, Joan Citoler, Ashley P. Mattey, Grayson J. Ford, Jack Sangster, Sebastian C. Cosgrove, Bruna Zucoloto da Costa, Rachel S. Heath, Thomas W. Thorpe, Yuqi Yu, Sabine L. Flitsch, Nicholas J. Turner. RetroBioCat Database: A Platform for Collaborative Curation and Automated Meta-Analysis of Biocatalysis Data. ACS Catalysis 2023, 13 (17) , 11771-11780. https://doi.org/10.1021/acscatal.3c01418
  18. Rocío Mercado, Steven M. Kearnes, Connor W. Coley. Data Sharing in Chemistry: Lessons Learned and a Case for Mandating Structured Reaction Data. Journal of Chemical Information and Modeling 2023, 63 (14) , 4253-4265. https://doi.org/10.1021/acs.jcim.3c00607
  19. Matisyahu S. Fogel, Kazunori Koide. Recent Progress on One-Pot Multisubstrate Screening. Organic Process Research & Development 2023, 27 (7) , 1235-1247. https://doi.org/10.1021/acs.oprd.3c00128
  20. Teresa M. Karl, Samir Bouayad-Gervais, Julian A. Hueffel, Theresa Sperger, Sebastian Wellig, Sherif J. Kaldas, Uladzislava Dabranskaya, Jas S. Ward, Kari Rissanen, Graham J. Tizzard, Franziska Schoenebeck. Machine Learning-Guided Development of Trialkylphosphine Ni(I) Dimers and Applications in Site-Selective Catalysis. Journal of the American Chemical Society 2023, 145 (28) , 15414-15424. https://doi.org/10.1021/jacs.3c03403
  21. Johanna Kleinekorte, Jonas Kleppich, Lorenz Fleitmann, Verena Beckert, Luise Blodau, André Bardow. APPROPRIATE Life Cycle Assessment: A PROcess-Specific, PRedictive Impact AssessmenT Method for Emerging Chemical Processes. ACS Sustainable Chemistry & Engineering 2023, 11 (25) , 9303-9319. https://doi.org/10.1021/acssuschemeng.2c07682
  22. Wei Liu, James Mulhearn, Bo Hao, Santiago Cañellas, Stefaan Last, José Enrique Gómez, Alexander Jones, Alexander De Vera, Kiran Kumar, Raquel Rodríguez, Lars Van Eynde, Iulia I. Strambeanu, Scott E. Wolkenberg. Enabling Deoxygenative C(sp2)-C(sp3) Cross-Coupling for Parallel Medicinal Chemistry. ACS Medicinal Chemistry Letters 2023, 14 (6) , 853-859. https://doi.org/10.1021/acsmedchemlett.3c00118
  23. Andrzej M. Żurański, Shivaani S. Gandhi, Abigail G. Doyle. A Machine Learning Approach to Model Interaction Effects: Development and Application to Alcohol Deoxyfluorination. Journal of the American Chemical Society 2023, 145 (14) , 7898-7909. https://doi.org/10.1021/jacs.2c13093
  24. Fernando Jaume-Santero, Alban Bornet, Alain Valery, Nona Naderi, David Vicente Alvarez, Dimitrios Proios, Anthony Yazdani, Colin Bournez, Thomas Fessard, Douglas Teodoro. Transformer Performance for Chemical Reactions: Analysis of Different Predictive and Evaluation Scenarios. Journal of Chemical Information and Modeling 2023, 63 (7) , 1914-1924. https://doi.org/10.1021/acs.jcim.2c01407
  25. Daniel S. Wigh, Matthieu Tissot, Patrick Pasau, Jonathan M. Goodman, Alexei A. Lapkin. Quantitative In Silico Prediction of the Rate of Protodeboronation by a Mechanistic Density Functional Theory-Aided Algorithm. The Journal of Physical Chemistry A 2023, 127 (11) , 2628-2636. https://doi.org/10.1021/acs.jpca.2c08250
  26. Connor J. Taylor, Alexander Pomberger, Kobi C. Felton, Rachel Grainger, Magda Barecka, Thomas W. Chamberlain, Richard A. Bourne, Christopher N. Johnson, Alexei A. Lapkin. A Brief Introduction to Chemical Reaction Optimization. Chemical Reviews 2023, 123 (6) , 3089-3126. https://doi.org/10.1021/acs.chemrev.2c00798
  27. Christopher Karpovich, Elton Pan, Zach Jensen, Elsa Olivetti. Interpretable Machine Learning Enabled Inorganic Reaction Classification and Synthesis Condition Prediction. Chemistry of Materials 2023, 35 (3) , 1062-1079. https://doi.org/10.1021/acs.chemmater.2c03010
  28. Martin Fitzner, Georg Wuitschik, Raffael Koller, Jean-Michel Adam, Torsten Schindler. Machine Learning C–N Couplings: Obstacles for a General-Purpose Reaction Yield Prediction. ACS Omega 2023, 8 (3) , 3017-3025. https://doi.org/10.1021/acsomega.2c05546
  29. Jose Raul Montero BastidasAbdellatif El MarrouniMaria Irina Chiriac, Thomas Struble, Dipannita Kalyani. ACCELERATING DRUG DISCOVERY BY HIGH-THROUGHPUT EXPERIMENTATION. , 443-463. https://doi.org/10.1021/mc-2022-vol57.ch18
  30. Youngchun Kwon, Dongseon Lee, Jin Woo Kim, Youn-Suk Choi, Sun Kim. Exploring Optimal Reaction Conditions Guided by Graph Neural Networks and Bayesian Optimization. ACS Omega 2022, 7 (49) , 44939-44950. https://doi.org/10.1021/acsomega.2c05165
  31. Youngchun Kwon, Sun Kim, Youn-Suk Choi, Seokho Kang. Generative Modeling to Predict Multiple Suitable Conditions for Chemical Reactions. Journal of Chemical Information and Modeling 2022, 62 (23) , 5952-5960. https://doi.org/10.1021/acs.jcim.2c01085
  32. Austin M. Mroz, Victor Posligua, Andrew Tarzia, Emma H. Wolpert, Kim E. Jelfs. Into the Unknown: How Computation Can Help Explore Uncharted Material Space. Journal of the American Chemical Society 2022, 144 (41) , 18730-18743. https://doi.org/10.1021/jacs.2c06833
  33. Liang Cao, Jianping Su, Yixiu Wang, Yankai Cao, Lim C. Siang, Jin Li, Jack Nicholas Saddler, Bhushan Gopaluni. Causal Discovery Based on Observational Data and Process Knowledge in Industrial Processes. Industrial & Engineering Chemistry Research 2022, 61 (38) , 14272-14283. https://doi.org/10.1021/acs.iecr.2c01326
  34. Jike Wang, Xiaorui Wang, Huiyong Sun, Mingyang Wang, Yundian Zeng, Dejun Jiang, Zhenxing Wu, Zeyi Liu, Ben Liao, Xiaojun Yao, Chang-Yu Hsieh, Dongsheng Cao, Xi Chen, Tingjun Hou. ChemistGA: A Chemical Synthesizable Accessible Molecular Generation Algorithm for Real-World Drug Discovery. Journal of Medicinal Chemistry 2022, 65 (18) , 12482-12496. https://doi.org/10.1021/acs.jmedchem.2c01179
  35. Lixue Cheng, Jiace Sun, Thomas F. Miller, III. Accurate Molecular-Orbital-Based Machine Learning Energies via Unsupervised Clustering of Chemical Space. Journal of Chemical Theory and Computation 2022, 18 (8) , 4826-4835. https://doi.org/10.1021/acs.jctc.2c00396
  36. Uschi Dolfus, Hans Briem, Matthias Rarey. Synthesis-Aware Generation of Structural Analogues. Journal of Chemical Information and Modeling 2022, 62 (15) , 3565-3576. https://doi.org/10.1021/acs.jcim.2c00246
  37. Li-Tao Zhu, Xi-Zhong Chen, Bo Ouyang, Wei-Cheng Yan, He Lei, Zhe Chen, Zheng-Hong Luo. Review of Machine Learning for Hydrodynamics, Transport, and Reactions in Multiphase Flows and Reactors. Industrial & Engineering Chemistry Research 2022, 61 (28) , 9901-9949. https://doi.org/10.1021/acs.iecr.2c01036
  38. Kevin A. Spiekermann, Lagnajit Pattanaik, William H. Green. Fast Predictions of Reaction Barrier Heights: Toward Coupled-Cluster Accuracy. The Journal of Physical Chemistry A 2022, 126 (25) , 3976-3986. https://doi.org/10.1021/acs.jpca.2c02614
  39. Zhiqing Xu, Radhakrishnan Mahadevan. Efficient Enumeration of Branched Novel Biochemical Pathways Using a Probabilistic Technique. Industrial & Engineering Chemistry Research 2022, 61 (25) , 8645-8657. https://doi.org/10.1021/acs.iecr.1c02211
  40. Hanyu Gao, Li-Tao Zhu, Zheng-Hong Luo, Marco A. Fraga, I-Ming Hsing. Machine Learning and Data Science in Chemical Engineering. Industrial & Engineering Chemistry Research 2022, 61 (24) , 8357-8358. https://doi.org/10.1021/acs.iecr.2c01788
  41. Alfredo Pereira, Camilo Albornoz, Oleksandra S. Trofymchuk. Data-Driven Analysis of Reactions Catalyzed by [CoCp*(CO)I2]. Organometallics 2022, 41 (10) , 1158-1166. https://doi.org/10.1021/acs.organomet.2c00051
  42. Jiang Guo, A. Santiago Ibanez-Lopez, Hanyu Gao, Victor Quach, Connor W. Coley, Klavs F. Jensen, Regina Barzilay. Automated Chemical Reaction Extraction from Scientific Literature. Journal of Chemical Information and Modeling 2022, 62 (9) , 2035-2045. https://doi.org/10.1021/acs.jcim.1c00284
  43. Alexe L. Haywood, Joseph Redshaw, Magnus W. D. Hanson-Heine, Adam Taylor, Alex Brown, Andrew M. Mason, Thomas Gärtner, Jonathan D. Hirst. Kernel Methods for Predicting Yields of Chemical Reactions. Journal of Chemical Information and Modeling 2022, 62 (9) , 2077-2092. https://doi.org/10.1021/acs.jcim.1c00699
  44. Jieyu Lu, Yingkai Zhang. Unified Deep Learning Model for Multitask Reaction Predictions with Explanation. Journal of Chemical Information and Modeling 2022, 62 (6) , 1376-1387. https://doi.org/10.1021/acs.jcim.1c01467
  45. Wiktor Beker, Rafał Roszak, Agnieszka Wołos, Nicholas H. Angello, Vandana Rathore, Martin D. Burke, Bartosz A. Grzybowski. Machine Learning May Sometimes Simply Capture Literature Popularity Trends: A Case Study of Heterocyclic Suzuki–Miyaura Coupling. Journal of the American Chemical Society 2022, 144 (11) , 4819-4827. https://doi.org/10.1021/jacs.1c12005
  46. Hyunsoo Park, Yeonghun Kang, Wonyoung Choe, Jihan Kim. Mining Insights on Metal–Organic Framework Synthesis from Scientific Literature Texts. Journal of Chemical Information and Modeling 2022, 62 (5) , 1190-1198. https://doi.org/10.1021/acs.jcim.1c01297
  47. Senthil M. Arumugam, Dalwinder Singh, Sangeeta Mahala, Bhawana Devi, Sandeep Kumar, Sunaina Jakhu, Sasikumar Elumalai. MgO/CaO Nanocomposite Facilitates Economical Production of d-Fructose and d-Allulose Using Glucose and Its Response Prediction Using a DNN Model. Industrial & Engineering Chemistry Research 2022, 61 (6) , 2524-2537. https://doi.org/10.1021/acs.iecr.1c04631
  48. Eisuke Sato, Mayu Fujii, Hiroki Tanaka, Koichi Mitsudo, Masaru Kondo, Shinobu Takizawa, Hiroaki Sasai, Takeshi Washio, Kazunori Ishikawa, Seiji Suga. Application of an Electrochemical Microflow Reactor for Cyanosilylation: Machine Learning-Assisted Exploration of Suitable Reaction Conditions for Semi-Large-Scale Synthesis. The Journal of Organic Chemistry 2021, 86 (22) , 16035-16044. https://doi.org/10.1021/acs.joc.1c01242
  49. Steven M. Kearnes, Michael R. Maser, Michael Wleklinski, Anton Kast, Abigail G. Doyle, Spencer D. Dreher, Joel M. Hawkins, Klavs F. Jensen, Connor W. Coley. The Open Reaction Database. Journal of the American Chemical Society 2021, 143 (45) , 18820-18826. https://doi.org/10.1021/jacs.1c09820
  50. Samuel Boobier, Yufeng Liu, Krishna Sharma, David R. J. Hose, A. John Blacker, Nikil Kapur, Bao N. Nguyen. Predicting Solvent-Dependent Nucleophilicity Parameter with a Causal Structure Property Relationship. Journal of Chemical Information and Modeling 2021, 61 (10) , 4890-4899. https://doi.org/10.1021/acs.jcim.1c00610
  51. Alexander J. S. Hammer, Artem I. Leonov, Nicola L. Bell, Leroy Cronin. Chemputation and the Standardization of Chemical Informatics. JACS Au 2021, 1 (10) , 1572-1587. https://doi.org/10.1021/jacsau.1c00303
  52. Kirill Karpov, Artem Mitrofanov, Vadim Korolev, Valery Tkachenko. Size Doesn’t Matter: Predicting Physico- or Biochemical Properties Based on Dozens of Molecules. The Journal of Physical Chemistry Letters 2021, 12 (38) , 9213-9219. https://doi.org/10.1021/acs.jpclett.1c02477
  53. Markus Meuwly. Machine Learning for Chemical Reactions. Chemical Reviews 2021, 121 (16) , 10218-10239. https://doi.org/10.1021/acs.chemrev.1c00033
  54. Udit Gupta, Dionisios G. Vlachos. Learning Chemistry of Complex Reaction Systems via a Python First-Principles Reaction Rule Stencil (pReSt) Generator. Journal of Chemical Information and Modeling 2021, 61 (7) , 3431-3441. https://doi.org/10.1021/acs.jcim.1c00297
  55. Joonyoung F. Joung, Minhi Han, Jinhyo Hwang, Minseok Jeong, Dong Hoon Choi, Sungnam Park. Deep Learning Optical Spectroscopy Based on Experimental Database: Potential Applications to Molecular Design. JACS Au 2021, 1 (4) , 427-438. https://doi.org/10.1021/jacsau.1c00035
  56. Andrzej M. Żurański, Jesus I. Martinez Alvarado, Benjamin J. Shields, Abigail G. Doyle. Predicting Reaction Yields via Supervised Learning. Accounts of Chemical Research 2021, 54 (8) , 1856-1865. https://doi.org/10.1021/acs.accounts.0c00770
  57. Hanyu Gao, Jean Pauphilet, Thomas J. Struble, Connor W. Coley, Klavs F. Jensen. Direct Optimization across Computer-Generated Reaction Networks Balances Materials Use and Feasibility of Synthesis Plans for Molecule Libraries. Journal of Chemical Information and Modeling 2021, 61 (1) , 493-504. https://doi.org/10.1021/acs.jcim.0c01032
  58. Michael R. Maser, Alexander Y. Cui, Serim Ryou, Travis J. DeLano, Yisong Yue, Sarah E. Reisman. Multilabel Classification Models for the Prediction of Cross-Coupling Reaction Conditions. Journal of Chemical Information and Modeling 2021, 61 (1) , 156-166. https://doi.org/10.1021/acs.jcim.0c01234
  59. Somesh Mohapatra, Nina Hartrampf, Mackenzie Poskus, Andrei Loas, Rafael Gómez-Bombarelli, Bradley L. Pentelute. Deep Learning for Prediction and Optimization of Fast-Flow Peptide Synthesis. ACS Central Science 2020, 6 (12) , 2277-2286. https://doi.org/10.1021/acscentsci.0c00979
  60. Andrew F. Zahrt, Jeremy J. Henle, Scott E. Denmark. Cautionary Guidelines for Machine Learning Studies with Combinatorial Datasets. ACS Combinatorial Science 2020, 22 (11) , 586-591. https://doi.org/10.1021/acscombsci.0c00118
  61. Krupal P. Jethava, Jonathan Fine, Yingqi Chen, Ahad Hossain, Gaurav Chopra. Accelerated Reactivity Mechanism and Interpretable Machine Learning Model of N-Sulfonylimines toward Fast Multicomponent Reactions. Organic Letters 2020, 22 (21) , 8480-8486. https://doi.org/10.1021/acs.orglett.0c03083
  62. Evan Komp, Stéphanie Valleau. Machine Learning Quantum Reaction Rate Constants. The Journal of Physical Chemistry A 2020, 124 (41) , 8607-8613. https://doi.org/10.1021/acs.jpca.0c05992
  63. Thomas J. Struble, Juan C. Alvarez, Scott P. Brown, Milan Chytil, Justin Cisar, Renee L. DesJarlais, Ola Engkvist, Scott A. Frank, Daniel R. Greve, Daniel J. Griffin, Xinjun Hou, Jeffrey W. Johannes, Constantine Kreatsoulas, Brian Lahue, Miriam Mathea, Georg Mogk, Christos A. Nicolaou, Andrew D. Palmer, Daniel J. Price, Richard I. Robinson, Sebastian Salentin, Li Xing, Tommi Jaakkola, William. H. Green, Regina Barzilay, Connor W. Coley, Klavs F. Jensen. Current and Future Roles of Artificial Intelligence in Medicinal Chemistry Synthesis. Journal of Medicinal Chemistry 2020, 63 (16) , 8667-8682. https://doi.org/10.1021/acs.jmedchem.9b02120
  64. Steven L. Rohall, Lydia Auch, Jonathan Gable, Jacob Gora, Johanna Jansen, Yipin Lu, Eric Martin, Margaret Pancost-Heidebrecht, Bill Shirley, Nikolaus Stiefl, Mika Lindvall. An Artificial Intelligence Approach to Proactively Inspire Drug Discovery with Recommendations. Journal of Medicinal Chemistry 2020, 63 (16) , 8824-8834. https://doi.org/10.1021/acs.jmedchem.9b02130
  65. Edward Kim, Zach Jensen, Alexander van Grootel, Kevin Huang, Matthew Staib, Sheshera Mysore, Haw-Shiuan Chang, Emma Strubell, Andrew McCallum, Stefanie Jegelka, Elsa Olivetti. Inorganic Materials Synthesis Planning with Literature-Trained Neural Networks. Journal of Chemical Information and Modeling 2020, 60 (3) , 1194-1201. https://doi.org/10.1021/acs.jcim.9b00995
  66. Manuel Moliner, Yuriy Román-Leshkov, Avelino Corma. Machine Learning Applied to Zeolite Synthesis: The Missing Link for Realizing High-Throughput Discovery. Accounts of Chemical Research 2019, 52 (10) , 2971-2980. https://doi.org/10.1021/acs.accounts.9b00399
  67. Tzyy-Shyang Lin, Connor W. Coley, Hidenobu Mochigase, Haley K. Beech, Wencong Wang, Zi Wang, Eliot Woods, Stephen L. Craig, Jeremiah A. Johnson, Julia A. Kalow, Klavs F. Jensen, Bradley D. Olsen. BigSMILES: A Structurally-Based Line Notation for Describing Macromolecules. ACS Central Science 2019, 5 (9) , 1523-1531. https://doi.org/10.1021/acscentsci.9b00476
  68. Eric Walker, Joshua Kammeraad, Jonathan Goetz, Michael T. Robo, Ambuj Tewari, Paul M. Zimmerman. Learning To Predict Reaction Conditions: Relationships between Solvent, Molecular Structure, and Catalyst. Journal of Chemical Information and Modeling 2019, 59 (9) , 3645-3654. https://doi.org/10.1021/acs.jcim.9b00313
  69. Kaushik Sivaramakrishnan, Anjana Puliyanda, Dereje Tamiru Tefera, Ajay Ganesh, Sushmitha Thirumalaivasan, Vinay Prasad. A Perspective on the Impact of Process Systems Engineering on Reaction Engineering. Industrial & Engineering Chemistry Research 2019, 58 (26) , 11149-11163. https://doi.org/10.1021/acs.iecr.9b00280
  70. John S. Schreck, Connor W. Coley, Kyle J. M. Bishop. Learning Retrosynthetic Planning through Simulated Experience. ACS Central Science 2019, 5 (6) , 970-981. https://doi.org/10.1021/acscentsci.9b00055
  71. Zach Jensen, Edward Kim, Soonhyoung Kwon, Terry Z. H. Gani, Yuriy Román-Leshkov, Manuel Moliner, Avelino Corma, Elsa Olivetti. A Machine Learning Approach to Zeolite Synthesis Enabled by Automatic Literature Data Extraction. ACS Central Science 2019, 5 (5) , 892-899. https://doi.org/10.1021/acscentsci.9b00193
  72. Shogo Nakamura, Nobuaki Yasuo, Masakazu Sekijima. Molecular optimization using a conditional transformer for reaction-aware compound exploration with reinforcement learning. Communications Chemistry 2025, 8 (1) https://doi.org/10.1038/s42004-025-01437-x
  73. Wenlong Wang, Chenyang Xu, Jian Du, Lei Zhang. Developing deep learning-based large-scale organic reaction classification model via sigma-profiles. Green Chemical Engineering 2025, 6 (2) , 181-192. https://doi.org/10.1016/j.gce.2024.06.003
  74. Wei Deng, Lijun Liu, Xiaohang Li, Yanyu Huang, Ming Hu, Yafang Zheng, Yuan Yin, Yan Huan, Shuxun Cui, Zhaoyan Sun, Jun Jiang, Xiaoniu Yang, Dapeng Wang. Machine‐Learning‐Enhanced Trial‐and‐Error for Efficient Optimization of Rubber Composites. Advanced Materials 2025, https://doi.org/10.1002/adma.202407763
  75. Amir Hossein Sheikhshoaei, Ali Khoshsima, Davood Zabihzadeh. Predicting the heat capacity of strontium-praseodymium oxysilicate SrPr4(SiO4)3O using machine learning, deep learning, and hybrid models. Chemical Thermodynamics and Thermal Analysis 2025, 17 , 100154. https://doi.org/10.1016/j.ctta.2024.100154
  76. Ella Gale, Leo Lobski, Fabio Zanasi. A categorical model for organic chemistry. Theoretical Computer Science 2025, 1032 , 115084. https://doi.org/10.1016/j.tcs.2025.115084
  77. Junu Kim, Kozue Okamura, Mohamed Rami Gaddem, Yusuke Hayashi, Sara Badr, Hirokazu Sugiyama. Impact of modeling and simulation on pharmaceutical process development. Current Opinion in Chemical Engineering 2025, 47 , 101093. https://doi.org/10.1016/j.coche.2025.101093
  78. Eunjae Shim, Ambuj Tewari, Tim Cernak, Paul M. Zimmerman. Recommending reaction conditions with label ranking. Chemical Science 2025, 16 (9) , 4109-4118. https://doi.org/10.1039/D4SC06728B
  79. Qilei Liu, Haitao Mao, Lu Wang, Lei Zhang. Hunting for Better Aromatic Chemicals with AI Techniques. 2025, 23-77. https://doi.org/10.1002/9783527845491.ch2
  80. Emil I. Jaffal, Sangjoon Lee, Danila Shiryaev, Alex Vtorov, Nikhil Kumar Barua, Holger Kleinke, Anton O. Oliynyk. Composition and structure analyzer/featurizer for explainable machine-learning models to predict solid state structures. Digital Discovery 2025, 4 (2) , 548-560. https://doi.org/10.1039/D4DD00332B
  81. Arunangshu Das, Anita Verma, Naba Hazarika. A comprehensive review on integration of cellular metabolic engineering and cell-free systems for microbial platforms. Process Biochemistry 2025, 149 , 222-236. https://doi.org/10.1016/j.procbio.2024.12.010
  82. Yuting Shang, Xiang Gao, Hongqin Wei, Zhengzheng Wang, Liqing Xi, Yantao Wang, Meijing Liu, Ying Feng, Juan Wang, Qingping Wu, Moutong Chen, Yu Ding. Advancements of prokaryotic Argonautes in molecular diagnostics and future perspectives. TrAC Trends in Analytical Chemistry 2025, 183 , 118122. https://doi.org/10.1016/j.trac.2024.118122
  83. Zexi Zhang, Zhanxiang Cai, Wenbin Zhang, Hua Lu, Mao Chen. Machine learning-assisted investigations toward polymer synthesis. Chinese Science Bulletin 2025, 70 (4-5) , 471-480. https://doi.org/10.1360/TB-2024-0800
  84. Xinghai Li, Zhisen Wu, Lijing Zhang, Shengyang Tao. Machine learning enables the prediction of amide bond synthesis based on small datasets. Acta Physico-Chimica Sinica 2025, 41 (2) , 100010. https://doi.org/10.3866/PKU.WHXB202309041
  85. P. Selvakumar, C. Preethi, P. Nehru, A. Saravanan, Abhijeet Das. AI in Green Chemistry Sustainable Manufacturing Processes. 2025, 297-318. https://doi.org/10.4018/979-8-3693-7483-2.ch011
  86. Weilong Hu, Enzhe Jing, Haoke Qiu, Zhao-Yan Sun. Discovering polyimides and their composites with targeted mechanical properties through explainable machine learning. Journal of Materials Informatics 2025, 5 (1) https://doi.org/10.20517/jmi.2024.59
  87. Zihan Wang, Kangjie Lin, Jianfeng Pei, Luhua Lai. Reacon: a template- and cluster-based framework for reaction condition prediction. Chemical Science 2025, 16 (2) , 854-866. https://doi.org/10.1039/D4SC05946H
  88. Inbal Lorena Eshel, Shahar Barkai, Sergio Barranco, Monica Hevia Perez-Temprano, Anat Milo. Probability Guided Chemical Reaction Scopes. 2025https://doi.org/10.2139/ssrn.5138219
  89. Zhenzhi Tan, Qi Yang, Sanzhong Luo. AI molecular catalysis: where are we now?. Organic Chemistry Frontiers 2025, 2 https://doi.org/10.1039/D4QO02363C
  90. Joseph C. Davies, Jonathan D. Hirst. Software Tools for Green and Sustainable Chemistry. 2025, 414-425. https://doi.org/10.1016/B978-0-443-15742-4.00049-1
  91. Yixin Wei, Leyu Shan, Tong Qiu, Diannan Lu, Zheng Liu. Machine learning-assisted retrosynthesis planning: Current status and future prospects. Chinese Journal of Chemical Engineering 2025, 77 , 273-292. https://doi.org/10.1016/j.cjche.2024.10.014
  92. C.F. Blanco, N. Pauliks, F. Donati, N. Engberg, J. Weber. Machine learning to support prospective life cycle assessment of emerging chemical technologies. Current Opinion in Green and Sustainable Chemistry 2024, 50 , 100979. https://doi.org/10.1016/j.cogsc.2024.100979
  93. Wen Luo, Yangyi Shen, Chengfan Fu, Xiao Feng, Qiang Huang. Exploring the CO2 conversion activated by the dielectric barrier discharge plasma assisted with photocatalyst via machine learning. Journal of Environmental Chemical Engineering 2024, 12 (6) , 114428. https://doi.org/10.1016/j.jece.2024.114428
  94. Buyong Ma, Yiguo Wang, Xingzi Li, Chang Shen, Hao Lin, Chenxi Du, Shanlin Yang, Ruoqing Zeng, Xuyang Tang, Jinglei Hu, Yukun Yang, Jingwen Wang, Jiawei Zhu, Xingqian Shan, Yu Zhang, Jiaqing Hu. Recent Advancements in the Application of Artificial Intelligence in Drug Molecular Generation and Synthesis Planning. Pharmaceutical Fronts 2024, 06 (04) , e394-e405. https://doi.org/10.1055/s-0044-1796647
  95. Lung-Yi Chen, Yi-Pei Li. Enhancing chemical synthesis: a two-stage deep neural network for predicting feasible reaction conditions. Journal of Cheminformatics 2024, 16 (1) https://doi.org/10.1186/s13321-024-00805-4
  96. Maarten R. Dobbelaere, István Lengyel, Christian V. Stevens, Kevin M. Van Geem. Rxn-INSIGHT: fast chemical reaction analysis using bond-electron matrices. Journal of Cheminformatics 2024, 16 (1) https://doi.org/10.1186/s13321-024-00834-z
  97. Lung-Yi Chen, Yi-Pei Li. AutoTemplate: enhancing chemical reaction datasets for machine learning applications in organic chemistry. Journal of Cheminformatics 2024, 16 (1) https://doi.org/10.1186/s13321-024-00869-2
  98. Tieu-Long Phan, Klaus Weinbauer, Thomas Gärtner, Daniel Merkle, Jakob L. Andersen, Rolf Fagerberg, Peter F. Stadler. Reaction rebalancing: a novel approach to curating reaction databases. Journal of Cheminformatics 2024, 16 (1) https://doi.org/10.1186/s13321-024-00875-4
  99. Saman Raza, Satya, Tahmeena Khan, Manisha Singh. AI Tools for Teaching-Learning Chemistry. 2024, 173-193. https://doi.org/10.2174/9789815305180124010011
  100. Xinyu Zhang, Daobin Mu, Shijie Lu, Yuanxing Zhang, Yuxiang Zhang, Zhuolin Yang, Zhikun Zhao, Borong Wu, Feng Wu. Ab Initio Design of Ni‐Rich Cathode Material with Assistance of Machine Learning for High Energy Lithium‐Ion Batteries. ENERGY & ENVIRONMENTAL MATERIALS 2024, 7 (6) https://doi.org/10.1002/eem2.12744
Load more citations
Download PDF
Go to ACS Central Science
Get e-Alerts
Get e-Alerts

ACS Central Science

Cite this: ACS Cent. Sci. 2018, 4, 11, 1465–1476
Click to copy citationCitation copied!
https://doi.org/10.1021/acscentsci.8b00357
Published November 16, 2018

Copyright © 2018 American Chemical Society. This publication is licensed under these Terms of Use.

Request reuse permissions

Article Views

45k

Altmetric

-

Citations

289
Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 1

    Figure 1. Change of the loss functions with the number of epochs (left figure, overall; right figure, chemical context and temperature).

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 2

    Figure 2. Relationship between the true temperature and the top-one predicted temperature (left panel), and predicted temperature if the predicted context matches the chemical context (right panel).

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 3

    Figure 3. Example of model predictions compared with recorded context (temperature rounded to the closest integer; black text represents the recorded conditions, and blue text represents the predicted conditions). (A) Nucleophilic epoxidation. (B) Deprotection of fluorenylmethyloxycarbonyl (Fmoc). (C) Luche reduction of eneone, TBS = tert-butyl(dimethyl)silyl. (D) Buchwald–Hartwig aryl amination, BINAP = 2,2′-bis(diphenylphosphino)-1,1′-binaphthyl. (E) Suzuki-Miyaura coupling, CyJohnPhos = (2-biphenyl)dicyclohexylphosphine. (F) Hoveyda–Grubbs cross metathesis.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 4

    Figure 4. Examples of the reactions with the fewest chemical elements matching the recorded context (temperature rounded to the nearest integer; black text represents the recorded conditions, and red text represents the predicted conditions). (A) Birch alkylation. (B) Hoveyda–Grubbs cross metathesis, TBS = tert-butyl(dimethyl)silyl. (C) Suzuki-Miyaura coupling. (D) Azide reduction.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 5

    Figure 5. Embedding of the most common 50 solvents projected onto a two-dimensional space using t-SNE. Solvents are naturally clustered into their corresponding classes (manually annotated).

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 6

    Figure 6. Embedding of the most common 50 reagents projected onto a two-dimensional space using t-SNE. Reagents are naturally clustered into their corresponding classes (manually annotated).

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 7

    Figure 7. Graphical representation of the neural-network model for context recommendation (“Hard Selection” refers to setting the value of the maximal element to one and zero for the rest, although the output of each classification task is a probability distribution).

    High Resolution Image
    Download MS PowerPoint Slide

  • References

    This article references 48 other publications.

    1. 1
      Robinson, R. LXIII.–A Synthesis of Tropinone. J. Chem. Soc., Trans. 1917, 111 (0), 762– 768,  DOI: 10.1039/CT9171100762
      1
      Synthesis of tropinone
      Robinson, Robert
      Journal of the Chemical Society, Transactions (1917), 111 (), 762-8CODEN: JCHTA3; ISSN:0368-1645.
      Of the derivs. of tropinone (I) none is so suitable for the identification small amts. of (I) as dipiperonylidenetropinone (A), from (I) and piperonal boiled in alc. with aq. KOH for 15 min., yellow needles, m. 214°, sparingly sol. in most org. solvents; yield, quant., gives a blue color with H2SO4; the acetate dissolves in H2O with a bright orange-yellow color, the soln. giving cryst. ppts. with HCl, HBr, HNO3, H2SO4, (CO2H)2, and picric acid; hydrochloride (B), yellow microneedles. As a test for (I) the formation of (A) may be carried out as follows: The soln. may first be acidified and evapd. in vacuo in order to remove volatile impurities such as acetone; alc., excess of piperonal and KOH are added, and the soln. heated on the H2O bath for a few min. and poured into Et2O. This is washed with H2O and shaken with a little dil. HCl, whereupon (B) crysts. (CH2CHO)2 was prepd. according to Harries (Ber. 34, 1494(1901)) from 7 g. of the oxime in 30 cc. H2O, 50 cc. H2O added, and the HNO2 and HNO3 neutralized with pptd. CaCO3. 5 g. acetone and 4.5 g. MeNH2 in 20 cc. H2O were added and the mixt. let stand 0.5 hr., after which the soln. was tested for (I) as above and a small amt. of (B) isolated. If the mixt. was allowed to stand overnight other products were formed which interfered with the identification of (I). A more certain result was obtained when the acetone was substituted by OC(CH2CO2Et)2, letting the mixt. stand overnight, distg. off the alc., boiling with dil. H2SO4 for 0.5 hr., evapg. in a high vacuum, making alk. with KOH, and distg. off the (I) with steam and identifying as above. Best results were obtained, however, when OC(CH2CO2H)2 was used and the mixt. let stand 50 hrs., filtered, acidified with HCl, and concd. in vacuo. The residue was made alk. and distd. with steam and the distillate acidified with HCl, concd. in vacuo, made alk., and extd. with Et2O. After drying rapidly over KOH and evapg. the residue crystd. partly, the crystals being identical with a sample of (I) prepd. by the oxidation of tropine, and an additional amt. of (B) being obtained from the non-cryst. portion. Final conditions for the synthesis have not as yet been worked out, but in another expt. a yield of 42% of the theory was obtained, based on the estimation of the amt. of aldehyde used by conversion of a portion into the diphenylhydrazone and estimation of (I) by conversion into (B). It is hoped that the reaction will prove of general application, and an attempt will be made to synthesize ψ-pelletierine (II) by condensing H2C(CH2CHO)2 with MeNH2 and an acetone deriv.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaC2sXhsFShsw%253D%253D&md5=860007da88776b311b4f0070c5e56e7f
    2. 2
      Corey, E. J.; Wipke, W. T. Computer Assisted Design of Complex Organic Syntheses. Science (Washington, DC, U. S.) 1969, 166 (3902), 178– 192,  DOI: 10.1126/science.166.3902.178
      2
      Computer-assisted design of complex organic syntheses
      Corey, Elias J.; Wipke, W. Todd
      Science (Washington, DC, United States) (1969), 166 (3902), 178-92CODEN: SCIEAS; ISSN:0036-8075.
      The application of digital computers to the generation of paths for the synthesis of complex org. mols. was discussed. Given the requisite computing and graphic communications hardware a set of programs of such power as to make computerized synthetic anal. an indispensable aid is possible.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaF1MXlt1eju78%253D&md5=67a17fda766d2a14be6255cd33bfaa65
    3. 3
      Cook, A.; Johnson, A. P.; Law, J.; Mirzazadeh, M.; Ravitz, O.; Simon, A. Computer-Aided Synthesis Design: 40 Years On. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2 (1), 79– 107,  DOI: 10.1002/wcms.61
      3
      Computer-aided synthesis design. 40 years on
      Cook, Anthony; Johnson, A. Peter; Law, James; Mirzazadeh, Mahdi; Ravitz, Orr; Simon, Aniko
      Wiley Interdisciplinary Reviews: Computational Molecular Science (2012), 2 (1), 79-107CODEN: WIRCAH; ISSN:1759-0884. (Wiley-Blackwell)
      A review. The discipline of retrosynthetic anal. is now just over 40 years old. From the earliest day, attempts were made to incorporate this approach into computer programs to test the extent in which chem. perception and synthetic thinking could be formalized. Despite pioneering research efforts, computer-aided synthetic anal. failed to achieve widespread routine use by chemists, which can be attributed in part to the difficulty of building the required high-quality retrosynthetic transform databases required for credible analyses. However, with the advent over the past 25 years of large comprehensive reaction databases, work on successfully automating the construction of reliable and comprehensive reaction rule databases is promising to revitalize research in this field. This review compares and contrasts the diverse approaches taken by selected programs in both the design and implementation of mol. feature perception and reaction rule representation, and the concepts of synthetic strategy selection, representation, and execution were reviewed. In particular, the current work on automating the construction of reliable and comprehensive synthetic rule sets from available reaction databases in newer programs such as ARChem were discussed. The authors argued that the progress achieved in this aspect paves the way to a deeper exploration of computer approaches to applying strategy and control in the synthesis problem.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhvFGls7g%253D&md5=a659f7b50039d0ba6de10e4e829b6a9e
    4. 4
      Warr, W. A. A Short Review of Chemical Reaction Database Systems, Computer-Aided Synthesis Design, Reaction Prediction and Synthetic Feasibility. Mol. Inf. 2014, 33 (6–7), 469– 476,  DOI: 10.1002/minf.201400052
      4
      A Short Review of Chemical Reaction Database Systems, Computer-Aided Synthesis Design, Reaction Prediction and Synthetic Feasibility
      Warr, Wendy A.
      Molecular Informatics (2014), 33 (6-7), 469-476CODEN: MIONBS; ISSN:1868-1743. (Wiley-VCH Verlag GmbH & Co. KGaA)
      This article is the text for a pedagogical lecture to be given at the Strasbourg Summer School in Chemoinformatics in June 2104. It covers a very wide range of reaction topics including structure and reaction representation, reaction centers, atom-to-atom mapping, reaction retrieval systems, computer-aided synthesis design, retrosynthesis, reaction prediction and synthetic feasibility. In the time available, the coverage of each topic can only be cursory; the main usefulness of this article to the research community is the extensive bibliog.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXovFCmur8%253D&md5=392354d1d71f1ad31f1087d56b9afd57
    5. 5
      Engkvist, O.; Norrby, P. O.; Selmi, N.; Lam, Y. H.; Peng, Z.; Sherer, E. C.; Amberg, W.; Erhard, T.; Smyth, L. A. Computational Prediction of Chemical Reactions: Current Status and Outlook. Drug Discovery Today 2018, 23 (6), 1203– 1218,  DOI: 10.1016/j.drudis.2018.02.014
      5
      Computational prediction of chemical reactions: current status and outlook
      Engkvist, Ola; Norrby, Per-Ola; Selmi, Nidhal; Lam, Yu-hong; Peng, Zhengwei; Sherer, Edward C.; Amberg, Willi; Erhard, Thomas; Smyth, Lynette A.
      Drug Discovery Today (2018), 23 (6), 1203-1218CODEN: DDTOFS; ISSN:1359-6446. (Elsevier Ltd.)
      A review. Over the past few decades, various computational methods have become increasingly important for discovering and developing novel drugs. Computational prediction of chem. reactions is a key part of an efficient drug discovery process. In this review, we discuss important parts of this field, with a focus on utilizing reaction data to build predictive models, the existing programs for synthesis prediction, and usage of quantum mechanics and mol. mechanics (QM/MM) to explore chem. reactions. We also outline potential future developments with an emphasis on pre-competitive collaboration opportunities.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXkslSitrk%253D&md5=527a5783d106ec915e6f7afbd82db210
    6. 6
      Coley, C. W.; Green, W. H.; Jensen, K. F. Machine Learning in Computer-Aided Synthesis Planning. Acc. Chem. Res. 2018, 51 (5), 1281– 1289,  DOI: 10.1021/acs.accounts.8b00087
      6
      Machine Learning in Computer-Aided Synthesis Planning
      Coley, Connor W.; Green, William H.; Jensen, Klavs F.
      Accounts of Chemical Research (2018), 51 (5), 1281-1289CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
      Computer-aided synthesis planning (CASP) is focused on the goal of accelerating the process by which chemists decide how to synthesize small mol. compds. The ideal CASP program would take a mol. structure as input and output a sorted list of detailed reaction schemes that each connect that target to purchasable starting materials via a series of chem. feasible reaction steps. Early work in this field relied on expert-crafted reaction rules and heuristics to describe possible retrosynthetic disconnections and selectivity rules but suffered from incompleteness, infeasible suggestions, and human bias. With the relatively recent availability of large reaction corpora (such as the United States Patent and Trademark Office (USPTO), Reaxys, and SciFinder databases), consisting of millions of tabulated reaction examples, it is now possible to construct and validate purely data-driven approaches to synthesis planning. As a result, synthesis planning has been opened to machine learning techniques, and the field is advancing rapidly. In this Account, we focus on two crit. aspects of CASP and recent machine learning approaches to both challenges. First, we discuss the problem of retrosynthetic planning, which requires a recommender system to propose synthetic disconnections starting from a target mol. We describe how the search strategy, necessary to overcome the exponential growth of the search space with increasing no. of reaction steps, can be assisted through a learned synthetic complexity metric. We also describe how the recursive expansion can be performed by a straightforward nearest neighbor model that makes clever use of reaction data to generate high quality retrosynthetic disconnections. Second, we discuss the problem of anticipating the products of chem. reactions, which can be used to validate proposed reactions in a computer-generated synthesis plan (i.e., reduce false positives) to increase the likelihood of exptl. success. While we introduce this task in the context of reaction validation, its utility extends to the prediction of side products and impurities, among other applications. We describe neural network-based approaches that we and others have developed for this forward prediction task that can be trained on previously published exptl. data. Machine learning and artificial intelligence have revolutionized a no. of disciplines, not limited to image recognition, dictation, translation, content recommendation, advertising, and autonomous driving. While there is a rich history of using machine learning for structure-activity models in chem., it is only now that it is being successfully applied more broadly to org. synthesis and synthesis design. As reported in this Account, machine learning is rapidly transforming CASP, but there are several remaining challenges and opportunities, many pertaining to the availability and standardization of both data and evaluation metrics, which must be addressed by the community at large.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXosFKhsb0%253D&md5=a1ea72c55942f3c0f0a99ab080f96899
    7. 7
      Goodman, J. M. Reaction Prediction and Synthesis Design. Appl. Chemoinformatics Achiev. Futur. Oppor. 2018, 86– 105,  DOI: 10.1002/9783527806539.ch4b
      There is no corresponding record for this reference.
    8. 8
      Reaxys. https://new.reaxys.com/ (accessed on Sept 28, 2017).
      There is no corresponding record for this reference.
    9. 9
      Lowe, D. M. Patent Reaction Extractor (v1.0); 2014.
      There is no corresponding record for this reference.
    10. 10
      Szymkuć, S.; Gajewska, E. P.; Klucznik, T.; Molga, K.; Dittwald, P.; Startek, M.; Bajczyk, M.; Grzybowski, B. A. Computer-Assisted Synthetic Planning: The End of the Beginning. Angew. Chem., Int. Ed. 2016, 55 (20), 5904– 5937,  DOI: 10.1002/anie.201506101
      10
      Computer-Assisted Synthetic Planning: The End of the Beginning
      Szymkuc, Sara; Gajewska, Ewa P.; Klucznik, Tomasz; Molga, Karol; Dittwald, Piotr; Startek, Michal; Bajczyk, Michal; Grzybowski, Bartosz A.
      Angewandte Chemie, International Edition (2016), 55 (20), 5904-5937CODEN: ACIEF5; ISSN:1433-7851. (Wiley-VCH Verlag GmbH & Co. KGaA)
      Exactly half a century has passed since the launch of the first documented research project (1965 Dendral) on computer-assisted org. synthesis. Many more programs were created in the 1970s and 1980s but the enthusiasm of these pioneering days had largely dissipated by the 2000s, and the challenge of teaching the computer how to plan org. syntheses earned itself the reputation of a "mission impossible". This is quite curious given that, in the meantime, computers have "learned" many other skills that had been considered exclusive domains of human intellect and creativity-for example, machines can nowadays play chess better than human world champions and they can compose classical music pleasant to the human ear. Although there have been no similar feats in org. synthesis, this Review argues that to concede defeat would be premature. Indeed, bringing together the combination of modern computational power and algorithms from graph/network theory, chem. rules (with full stereo- and regiochem.) coded in appropriate formats, and the elements of quantum mechanics, the machine can finally be "taught" how to plan syntheses of non-trivial org. mols. in a matter of seconds to minutes. The Review begins with an overview of some basic theor. concepts essential for the big-data anal. of chem. syntheses. It progresses to the problem of optimizing pathways involving known reactions. It culminates with discussion of algorithms that allow for a completely de novo and fully automated design of syntheses leading to relatively complex targets, including those that have not been made before. Of course, there are still things to be improved, but computers are finally becoming relevant and helpful to the practice of org.-synthetic planning. Paraphrasing Churchill's famous words after the Allies' first major victory over the Axis forces in Africa, it is not the end, it is not even the beginning of the end, but it is the end of the beginning for the computer-assisted synthesis planning. The machine is here to stay.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XlvVagsbs%253D&md5=f07bb3b25f9b87c549860d7d081ed28f
    11. 11
      Segler, M. H.S.; Preuss, M.; Waller, M. P. Learning to Plan Chemical Syntheses. 2017, arXiv:1708.04202. arXiv.org e-Print archive. https://arxiv.org/abs/1708.04202.
      There is no corresponding record for this reference.
    12. 12
      Law, J.; Zsoldos, Z.; Simon, A.; Reid, D.; Liu, Y.; Khew, S. Y.; Johnson, A. P.; Major, S.; Wade, R. A.; Ando, H. Y. Route Designer: A Retrosynthetic Analysis Tool Utilizing Automated Retrosynthetic Rule Generation. J. Chem. Inf. Model. 2009, 49 (3), 593– 602,  DOI: 10.1021/ci800228y
      12
      Route Designer: A Retrosynthetic Analysis Tool Utilizing Automated Retrosynthetic Rule Generation
      Law, James; Zsoldos, Zsolt; Simon, Aniko; Reid, Darryl; Liu, Yang; Khew, Sing Yoong; Johnson, A. Peter; Major, Sarah; Wade, Robert A.; Ando, Howard Y.
      Journal of Chemical Information and Modeling (2009), 49 (3), 593-602CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
      Route Designer, version 1.0, is a new retrosynthetic anal. package that generates complete synthetic routes for target mols. starting from readily available starting materials. Rules describing retrosynthetic transformations are automatically generated from reaction databases, which ensure that the rules can be easily updated to reflect the latest reaction literature. These rules are used to carry out an exhaustive retrosynthetic anal. of the target mol., in which heuristics are used to mitigate the combinatorial explosion. Proposed routes are prioritized by an empirical rating algorithm to present a diverse profile of the most promising solns. The program runs on a server with a web-based user interface. An overview of the system is presented together with examples that illustrate Route Designer's utility.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXhsFegtL8%253D&md5=ef949298d3c3201c70bcfc2af56218ea
    13. 13
      Liu, B.; Ramsundar, B.; Kawthekar, P.; Shi, J.; Gomes, J.; Luu Nguyen, Q.; Ho, S.; Sloane, J.; Wender, P.; Pande, V. Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models. ACS Cent. Sci. 2017, 3 (10), 1103– 1113,  DOI: 10.1021/acscentsci.7b00303
      13
      Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models
      Liu, Bowen; Ramsundar, Bharath; Kawthekar, Prasad; Shi, Jade; Gomes, Joseph; Nguyen, Quang Luu; Ho, Stephen; Sloane, Jack; Wender, Paul; Pande, Vijay
      ACS Central Science (2017), 3 (10), 1103-1113CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
      We describe a fully data driven model that learns to perform a retrosynthetic reaction prediction task, which is treated as a sequence-to-sequence mapping problem. The end-to-end trained model has an encoder-decoder architecture that consists of two recurrent neural networks, which has previously shown great success in solving other sequence-to-sequence prediction tasks such as machine translation. The model is trained on 50,000 exptl. reaction examples from the United States patent literature, which span 10 broad reaction types that are commonly used by medicinal chemists. We find that our model performs comparably with a rule-based expert system baseline model, and also overcomes certain limitations assocd. with rule-based expert systems and with any machine learning approach that contains a rule-based expert system component. Our model provides an important first step toward solving the challenging problem of computational retrosynthetic anal.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXhsVahu7fI&md5=61b6213efc544c8e24e4aa8b750e28d3
    14. 14
      Bøgevig, A.; Federsel, H.-J.; Huerta, F.; Hutchings, M. G.; Kraut, H.; Langer, T.; Löw, P.; Oppawsky, C.; Rein, T.; Saller, H. Software Tool as an Idea Generator for Synthesis Prediction. Org. Process Res. Dev. 2015, 19 (2), 357– 368,  DOI: 10.1021/op500373e
      There is no corresponding record for this reference.
    15. 15
      Coley, C. W.; Rogers, L.; Green, W. H.; Jensen, K. F. Computer-Assisted Retrosynthesis Based on Molecular Similarity. ACS Cent. Sci. 2017, 3 (12), 1237– 1245,  DOI: 10.1021/acscentsci.7b00355
      15
      Computer-Assisted Retrosynthesis Based on Molecular Similarity
      Coley, Connor W.; Rogers, Luke; Green, William H.; Jensen, Klavs F.
      ACS Central Science (2017), 3 (12), 1237-1245CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
      We demonstrate mol. similarity to be a surprisingly effective metric for proposing and ranking one-step retrosynthetic disconnections based on analogy to precedent reactions. The developed approach mimics the retrosynthetic strategy defined implicitly by a corpus of known reactions without the need to encode any chem. knowledge. Using 40000 reactions from the patent literature as a knowledge base, the recorded reactants are among the top 10 proposed precursors in 74.1% of 5000 test reactions, providing strong quant. support for our methodol. Extension of the one-step strategy to multistep pathway planning is demonstrated and discussed for two exemplary drug products.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXhvVSmurfN&md5=53ca1ab17142a856afce4ffb67e6acea
    16. 16
      Segler, M. H.S.; Preuss, M.; Waller, M. P. Planning Chemical Syntheses with Deep Neural Networks and Symbolic AI. Nature 2018, 555 (7698), 604– 610,  DOI: 10.1038/nature25978
      16
      Planning chemical syntheses with deep neural networks and symbolic AI
      Segler, Marwin H. S.; Preuss, Mike; Waller, Mark P.
      Nature (London, United Kingdom) (2018), 555 (7698), 604-610CODEN: NATUAS; ISSN:0028-0836. (Nature Research)
      To plan the syntheses of small org. mols., chemists use retrosynthesis, a problem-solving technique in which target mols. are recursively transformed into increasingly simpler precursors. Computer-aided retrosynthesis would be a valuable tool but at present it is slow and provides results of unsatisfactory quality. Here, we use Monte Carlo tree search and symbolic artificial intelligence (AI) to discover retrosynthetic routes. We combined Monte Carlo tree search with an expansion policy network that guides the search, and a filter network to pre-select the most promising retrosynthetic steps. These deep neural networks were trained on essentially all reactions ever published in org. chem. Our system solves for almost twice as many mols., thirty times faster than the traditional computer-aided search method, which is based on extd. rules and hand-designed heuristics. In a double-blind AB test, chemists on av. considered our computer-generated routes to be equiv. to reported literature routes.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXmsVGqt7c%253D&md5=400e9945ff83ffe2d12278aa4c562893
    17. 17
      Kayala, M. A.; Baldi, P. F. Learning to Predict Chemical Reactions. J. Chem. Inf. Model. 2011, 51 (9), 2209– 2222,  DOI: 10.1021/ci200207y
      17
      Learning to Predict Chemical Reactions
      Kayala, Matthew A.; Azencott, Chloe-Agathe; Chen, Jonathan H.; Baldi, Pierre
      Journal of Chemical Information and Modeling (2011), 51 (9), 2209-2222CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
      Being able to predict the course of arbitrary chem. reactions is essential to the theory and applications of org. chem. Approaches to the reaction prediction problems can be organized around three poles corresponding to: (1) phys. laws; (2) rule-based expert systems; and (3) inductive machine learning. Previous approaches at these poles, resp., are not high throughput, are not generalizable or scalable, and lack sufficient data and structure to be implemented. We propose a new approach to reaction prediction utilizing elements from each pole. Using a phys. inspired conceptualization, we describe single mechanistic reactions as interactions between coarse approxns. of MOs (MOs) and use topol. and physicochem. attributes as descriptors. Using an existing rule-based system (Reaction Explorer), we derive a restricted chem. data set consisting of 1630 full multistep reactions with 2358 distinct starting materials and intermediates, assocd. with 2989 productive mechanistic steps and 6.14 million unproductive mechanistic steps. And from machine learning, we pose identifying productive mechanistic steps as a statistical ranking, information retrieval problem: given a set of reactants and a description of conditions, learn a ranking model over potential filled-to-unfilled MO interactions such that the top-ranked mechanistic steps yield the major products. The machine learning implementation follows a two-stage approach, in which we first train atom level reactivity filters to prune 94.00% of nonproductive reactions with a 0.01% error rate. Then, we train an ensemble of ranking models on pairs of interacting MOs to learn a relative productivity function over mechanistic steps in a given system. Without the use of explicit transformation patterns, the ensemble perfectly ranks the productive mechanism at the top 89.05% of the time, rising to 99.86% of the time when the top four are considered. Furthermore, the system is generalizable, making reasonable predictions over reactants and conditions which the rule-based expert does not handle. A web interface to the machine learning based mechanistic reaction predictor is accessible through our chemoinformatics portal (http://cdb.ics.uci.edu) under the Toolkits section.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhtFSks7jP&md5=1b40eaa05deaeb36da0ede14bf8d29f9
    18. 18
      Kayala, M. A.; Baldi, P. ReactionPredictor: Prediction of Complex Chemical Reactions at the Mechanistic Level Using Machine Learning. J. Chem. Inf. Model. 2012, 52 (10), 2526– 2540,  DOI: 10.1021/ci3003039
      18
      ReactionPredictor: Prediction of Complex Chemical Reactions at the Mechanistic Level Using Machine Learning
      Kayala, Matthew A.; Baldi, Pierre
      Journal of Chemical Information and Modeling (2012), 52 (10), 2526-2540CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
      Proposing reasonable mechanisms and predicting the course of chem. reactions is important to the practice of org. chem. Approaches to reaction prediction have historically used obfuscating representations and manually encoded patterns or rules. Here we present ReactionPredictor, a machine learning approach to reaction prediction that models elementary, mechanistic reactions as interactions between approx. MOs (MOs). A training data set of productive reactions known to occur at reasonable rates and yields and verified by inclusion in the literature or textbooks is derived from an existing rule-based system and expanded upon with manual curation from graduate level textbooks. Using this training data set of complex polar, hypervalent, radical, and pericyclic reactions, a two-stage machine learning prediction framework is trained and validated. In the first stage, filtering models trained at the level of individual MOs are used to reduce the space of possible reactions to consider. In the second stage, ranking models over the filtered space of possible reactions are used to order the reactions such that the productive reactions are the top ranked. The resulting model, ReactionPredictor, perfectly ranks polar reactions 78.1% of the time and recovers all productive reactions 95.7% of the time when allowing for small nos. of errors. Pericyclic and radical reactions are perfectly ranked 85.8% and 77.0% of the time, resp., rising to >93% recovery for both reaction types with a small no. of allowed errors. Decisions about which of the polar, pericyclic, or radical reaction type ranking models to use can be made with >99% accuracy. Finally, for multistep reaction pathways, we implement the first mechanistic pathway predictor using constrained tree-search to discover a set of reasonable mechanistic steps from given reactants to given products. Webserver implementations of both the single step and pathway versions of ReactionPredictor are available via the chemoinformatics portal http://cdb.ics.uci.edu/.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhtlGjtrjF&md5=3507988197eb3c3400045d0bfc939ff1
    19. 19
      Segler, M. H.S.; Waller, M. P. Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction. Chem. - Eur. J. 2017, 23 (25), 5966– 5971,  DOI: 10.1002/chem.201605499
      19
      Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction
      Segler, Marwin H. S.; Waller, Mark P.
      Chemistry - A European Journal (2017), 23 (25), 5966-5971CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
      Reaction prediction and retrosynthesis are the cornerstones of org. chem. Rule-based expert systems have been the most widespread approach to computationally solve these two related challenges to date. However, reaction rules often fail because they ignore the mol. context, which leads to reactivity conflicts. Herein, we report that deep neural networks can learn to resolve reactivity conflicts and to prioritize the most suitable transformation rules. We show that by training our model on 3.5 million reactions taken from the collective published knowledge of the entire discipline of chem., our model exhibits a top10-accuracy of 95 % in retrosynthesis and 97 % for reaction prediction on a validation set of almost 1 million reactions.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXjtlynsrw%253D&md5=e4f689a132ea6ad8713fa2d3d9422c78
    20. 20
      Jin, W.; Coley, C. W.; Barzilay, R.; Jaakkola, T. Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network , 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA; 2017; pp 2604– 2613.
      There is no corresponding record for this reference.
    21. 21
      Coley, C. W.; Barzilay, R.; Jaakkola, T. S.; Green, W. H.; Jensen, K. F. Prediction of Organic Reaction Outcomes Using Machine Learning. ACS Cent. Sci. 2017, 3 (5), 434– 443,  DOI: 10.1021/acscentsci.7b00064
      21
      Prediction of Organic Reaction Outcomes Using Machine Learning
      Coley, Connor W.; Barzilay, Regina; Jaakkola, Tommi S.; Green, William H.; Jensen, Klavs F.
      ACS Central Science (2017), 3 (5), 434-443CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
      Computer assistance in synthesis design has existed for over 40 years, yet retrosynthesis planning software has struggled to achieve widespread adoption. One crit. challenge in developing high-quality pathway suggestions is that proposed reaction steps often fail when attempted in the lab., despite initially seeming viable. The true measure of success for any synthesis program is whether the predicted outcome matches what is obsd. exptl. We report a model framework for anticipating reaction outcomes that combines the traditional use of reaction templates with the flexibility in pattern recognition afforded by neural networks. Using 15 000 exptl. reaction records from granted United States patents, a model is trained to select the major (recorded) product by ranking a self-generated list of candidates where one candidate is known to be the major product. Candidate reactions are represented using a unique edit-based representation that emphasizes the fundamental transformation from reactants to products, rather than the constituent mols.' overall structures. In a 5-fold cross-validation, the trained model assigns the major product rank 1 in 71.8% of cases, rank ≤3 in 86.7% of cases, and rank ≤5 in 90.8% of cases.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXmtVyqtb0%253D&md5=cc3e8a50d8dd9d68e294a21a558200dc
    22. 22
      Schwaller, P.; Laino, T. “Found in Translation”: Predicting Outcomes of Complex Organic Chemistry Reactions Using Neural Sequence-to-Sequence Models. 2017, arXiv:1711.04810. arXiv.org e-Print archive. https://arxiv.org/abs/1711.04810.
      There is no corresponding record for this reference.
    23. 23
      Reizman, B. J.; Jensen, K. F. Simultaneous Solvent Screening and Reaction Optimization in Microliter Slugs. Chem. Commun. 2015, 51 (68), 13290– 13293,  DOI: 10.1039/C5CC03651H
      23
      Simultaneous solvent screening and reaction optimization in microliter slugs
      Reizman, Brandon J.; Jensen, Klavs F.
      Chemical Communications (Cambridge, United Kingdom) (2015), 51 (68), 13290-13293CODEN: CHCOFS; ISSN:1359-7345. (Royal Society of Chemistry)
      An automated, continuous flow droplet screening system is presented, enabling real-time simultaneous solvent and continuous variable optimization. An optimal design of expts. strategy is applied to the alkylation of 1,2-diaminocyclohexane in 16 μL droplets, with scale-up demonstrated. Anal. of segmented flow results suggests correlation of yield with solvent hydrogen bond basicity.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXht1SqtL7O&md5=a27eeadd9328ade293e218e0167c87b8
    24. 24
      Sans, V.; Porwol, L.; Dragone, V.; Cronin, L. A Self Optimizing Synthetic Organic Reactor System Using Real-Time In-Line NMR Spectroscopy. Chem. Sci. 2015, 6 (2), 1258– 1264,  DOI: 10.1039/C4SC03075C
      24
      A self optimizing synthetic organic reactor system using real-time in-line NMR spectroscopy
      Sans, Victor; Porwol, Luzian; Dragone, Vincenza; Cronin, Leroy
      Chemical Science (2015), 6 (2), 1258-1264CODEN: CSHCCN; ISSN:2041-6520. (Royal Society of Chemistry)
      A configurable platform for synthetic chem. incorporating an in-line bench-top NMR capable of monitoring and controlling org. reactions in real-time is discussed. The platform is controlled by a modular LabView software control system for hardware, NMR, data anal., and feedback optimization. Using this platform, real-time advanced structural characterization of reaction mixts., including 19F, 13C, DEPT, 2-dimensional NMR spectroscopy (COSY, HSQC, 19F-COSY), are reported for the first time. The potential of this technique was demonstrated by optimizing a catalytic org. reaction in real-time, showing its applicability to self-optimizing systems using criteria such as stereo-selectivity, multi-nuclear measurements, or 2-dimensional correlations.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXhvFeisb%252FO&md5=e3d6d3dbfb1ba77055defb4078800dc1
    25. 25
      Holmes, N.; Akien, G. R.; Savage, R. J.D.; Stanetty, C.; Baxendale, I. R.; Blacker, A. J.; Taylor, B. A.; Woodward, R. L.; Meadows, R. E.; Bourne, R. A. Reaction Chemistry & Engineering Online Quantitative Mass Spectrometry for the Reactors †. React. Chem. Eng. 2016, 1 (1), 96– 100,  DOI: 10.1039/C5RE00083A
      25
      Online quantitative mass spectrometry for the rapid adaptive optimisation of automated flow reactors
      Holmes, Nicholas; Akien, Geoffrey R.; Savage, Robert J. D.; Stanetty, Christian; Baxendale, Ian R.; Blacker, A. John; Taylor, Brian A.; Woodward, Robert L.; Meadows, Rebecca E.; Bourne, Richard A.
      Reaction Chemistry & Engineering (2016), 1 (1), 96-100CODEN: RCEEBW; ISSN:2058-9883. (Royal Society of Chemistry)
      An automated continuous reactor for the synthesis of org. compds., which uses online mass spectrometry (MS) for reaction monitoring and product quantification, is presented. Quant. and rapid MS monitoring was developed and calibrated using HPLC. The amidation of Me nicotinate with aq. MeNH2 was optimized using design of expts. and a self-optimization algorithm approach to produce >93% yield.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhtlSlt7zM&md5=a8755576a3a4e7c07c10d7f1f3f675f2
    26. 26
      Reizman, B. J.; Jensen, K. F. Feedback in Flow for Accelerated Reaction Development. Acc. Chem. Res. 2016, 49 (9), 1786– 1796,  DOI: 10.1021/acs.accounts.6b00261
      26
      Feedback in Flow for Accelerated Reaction Development
      Reizman, Brandon J.; Jensen, Klavs F.
      Accounts of Chemical Research (2016), 49 (9), 1786-1796CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
      The pharmaceutical industry is investing in continuous flow and high-throughput experimentation as tools for rapid process development accelerated scale-up. Coupled with automation, these technologies offer the potential for comprehensive reaction characterization and optimization, but with the cost of conducting exhaustive multifactor screens. Automated feedback in flow offers researchers an alternative strategy for efficient characterization of reactions based on the use of continuous technol. to control chem. reaction conditions and optimize in lieu of screening. Optimization with feedback allows expts. to be conducted where the most information can be gained from the chem., enabling product yields to be maximized and kinetic models to be generated while the total no. of expts. is minimized. This account opens by reviewing select examples of feedback optimization in flow and applications to chem. research. Systems in the literature are classified into (i) deterministic black box optimization systems that do not model the reaction system and are therefore limited in the utility of results for scale-up, (ii) deterministic model-based optimization systems from which reaction kinetics and/or mechanisms can be automatically evaluated, and (iii) stochastic systems. Though diverse in application, flow feedback systems have predominantly focused upon the optimization of continuous variables, i.e., variables such as time, temp., and concn. that can be ramped from one expt. to the next. Unfortunately, this implies that the screening of discrete variables such as catalyst, ligand, or solvent generally does not factor into automated flow optimization, resulting in incomplete process knowledge. Herein, a system and strategy is presented developed for optimizing discrete and continuous variables of a chem. reaction simultaneously. The approach couples automated feedback with high-throughput reaction screening in droplet flow microfluidics. This account details the system configuration for on-demand creation of sub-20 μL droplets with interchangeable reagents and catalysts. These droplets are reacted in a fully automated microfluidic system and analyzed online by LC/MS. Feeding back from the online anal. results, a design of expts. (DoE)-based adaptive response surface algorithm is employed that deductively removes candidate reagents from the optimization as optimal reaction conditions are refined, leading to rapid convergence. Using the automated optimization platform, case studies are presented for solvent selection in a competitive alkylation chem. and for catalyst-ligand selection in heteroarom. Suzuki-Miyaura cross-coupling chemistries. For the monoalkylation of trans-1,2-diaminocyclohexane, polar aprotic solvents at moderate temps. are shown to be favorable, with optimality accurately identified with DMSO as the solvent in 67 expts. For Suzuki-Miyaura cross-couplings, the optimality of precatalysts and continuous variable conditions are obsd. to change in accordance with the coupling reagents, providing insights into catalyst behavior in the context of the reaction mechanism. Future opportunities in automated reaction development include the incorporation of chemoinformatics for faster anal. and machine-learning algorithms to guide and optimize the synthesis. Adoption of this technol. stands to reduce graduate student and postdoc time on routine tasks in the lab., while feeding back knowledge used to guide new research directions. Moreover, the application of this technol. in industry promises to lessen the cost and time assocd. with advancing pharmaceutical mols. through development and scale-up.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28Xhtlaru7%252FI&md5=ddaef4153972b8b045df12015e5b5d14
    27. 27
      Baumgartner, L. M.; Coley, C. W.; Reizman, B. J.; Gao, K. W.; Jensen, K. F. Optimum Catalyst Selection over Continuous and Discrete Process Variables with a Single Droplet Microfluidic Reaction Platform. React. Chem. Eng. 2018, 3 (3), 301– 311,  DOI: 10.1039/C8RE00032H
      27
      Optimum catalyst selection over continuous and discrete process variables with a single droplet microfluidic reaction platform
      Baumgartner, Lorenz M.; Coley, Connor W.; Reizman, Brandon J.; Gao, Kevin W.; Jensen, Klavs F.
      Reaction Chemistry & Engineering (2018), 3 (3), 301-311CODEN: RCEEBW; ISSN:2058-9883. (Royal Society of Chemistry)
      A mixed-integer nonlinear program (MINLP) algorithm to optimize catalyst turnover no. (TON) and product yield by simultaneously modulating discrete variables-catalyst types-and continuous variables-temp., residence time, and catalyst loading-was implemented and validated. Several simulated case studies, with and without random measurement error, demonstrate the algorithm's robustness in finding optimal conditions in the presence of side reactions and other complicating nonlinearities. This algorithm was applied to the real-time optimization of a Suzuki-Miyaura cross-coupling reaction in an automated microfluidic reaction platform comprising a liq. handler, an oscillatory flow reactor, and an online LC/MS. The algorithm, based on a combination of branch and bound and adaptive response surface methods, identified exptl. conditions that maximize TON subject to a yield constraint from a pool of eight catalyst candidates in just 60 expts., considerably fewer than a previous version of the algorithm.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXntlCntbo%253D&md5=a77e6dec5646d61ecc5eb42d04786fb4
    28. 28
      Kamlet, M. J.; Abboud, J. L.M.; Abraham, M. H.; Taft, R. W. Linear Solvation Energy Relationships. 23. A Comprehensive Collection of the Solvatochromic Parameters,.Pi.*,.Alpha., and.Beta., and Some Methods for Simplifying the Generalized Solvatochromic Equation. J. Org. Chem. 1983, 48 (17), 2877– 2887,  DOI: 10.1021/jo00165a018
      28
      Linear solvation energy relationships. 23. A comprehensive collection of the solvatochromic parameters, π*, α, and β, and some methods for simplifying the generalized solvatochromic equation
      Kamlet, Mortimer J.; Abboud, Jose Luis M.; Abraham, Michael H.; Taft, R. W.
      Journal of Organic Chemistry (1983), 48 (17), 2877-87CODEN: JOCEAH; ISSN:0022-3263.
      A generalized equation for linear solvation energy relations or complexation energy relations is developed which involves 6 terms: π* (a solvent dipolarity-polarizability term), α (solvent hydrogen-bond acceptor term), β (solvent hydrogen-bond donor term), δ (solvent polarizability correction term), δH (Hildebrand soly. parameter), and ξ. This equation is reduced to a more manageable form by a judicious choice of solvents and reactants or indicators. One-, two- or three-parameter LFER involving different combinations of the above parameters and various types of physicochem. properties are obsd. A comprehensive collection of π*, α, and β for 217 solvents is presented.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaL3sXkvVOgsbc%253D&md5=99b60c3cc817d5508b28d85f3ad6c8a1
    29. 29
      Struebing, H.; Ganase, Z.; Karamertzanis, P. G.; Siougkrou, E.; Haycock, P.; Piccione, P. M.; Armstrong, A.; Galindo, A.; Adjiman, C. S. Computer-Aided Molecular Design of Solvents for Accelerated Reaction Kinetics. Nat. Chem. 2013, 5 (11), 952– 957,  DOI: 10.1038/nchem.1755
      29
      Computer-aided molecular design of solvents for accelerated reaction kinetics
      Struebing, Heiko; Ganase, Zara; Karamertzanis, Panagiotis G.; Siougkrou, Eirini; Haycock, Peter; Piccione, Patrick M.; Armstrong, Alan; Galindo, Amparo; Adjiman, Claire S.
      Nature Chemistry (2013), 5 (11), 952-957CODEN: NCAHBB; ISSN:1755-4330. (Nature Publishing Group)
      Solvents can significantly alter the rates and selectivity of liq.-phase org. reactions, often hindering the development of new synthetic routes or, if chosen wisely, facilitating routes by improving rates and selectivities. To address this challenge, a systematic methodol. is proposed that quickly identifies improved reaction solvents by combining quantum mech. computations of the reaction rate const. in a few solvents with a computer-aided mol. design (CAMD) procedure. The approach allows the identification of a high-performance solvent within a very large set of possible mols. The validity of the authors' CAMD approach is demonstrated through application to a classical nucleophilic substitution reaction for the study of solvent effects, the Menschutkin reaction. The results were validated successfully by in situ kinetic expts. A space of 1,341 solvents was explored in silico, but required quantum-mech. calcns. of the rate const. in only nine solvents, and uncovered a solvent that increases the rate const. by 40%.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3sXhsV2jsrbP&md5=2fcd40cb79d2e6db80227791a628da0b
    30. 30
      Marcou, G.; Aires De Sousa, J.; Latino, D. A.R.S.; De Luca, A.; Horvath, D.; Rietsch, V.; Varnek, A. Expert System for Predicting Reaction Conditions: The Michael Reaction Case. J. Chem. Inf. Model. 2015, 55 (2), 239– 250,  DOI: 10.1021/ci500698a
      30
      Expert System for Predicting Reaction Conditions: The Michael Reaction Case
      Marcou, G.; Aires de Sousa, J.; Latino, D. A. R. S.; de Luca, A.; Horvath, D.; Rietsch, V.; Varnek, A.
      Journal of Chemical Information and Modeling (2015), 55 (2), 239-250CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
      A generic chem. transformation may often be achieved under various synthetic conditions. However, for any specific reagents, only one or a few among the reported synthetic protocols may be successful. For example, Michael β-addn. reactions may proceed under different choices of solvent (e.g., hydrophobic, aprotic polar, protic) and catalyst (e.g., Bronsted acid, Lewis acid, Lewis base, etc.). Chemoinformatics methods could be efficiently used to establish a relationship between the reagent structures and the required reaction conditions, which would allow synthetic chemists to waste less time and resources in trying out various protocols in search for the appropriate one. In order to address this problem, a no. of 2-classes classification models have been built on a set of 198 Michael reactions retrieved from literature. Trained models discriminate between processes that are compatible and resp. processes not feasible under a specific reaction condition option (feasible or not with a Lewis acid catalyst, feasible or not in hydrophobic solvent, etc.). Eight distinct models were built to decide the compatibility of a Michael addn. process with each considered reaction condition option, while a ninth model was aimed to predict whether the assumed Michael addn. is feasible at all. Different machine-learning methods (Support Vector Machine, Naive Bayes, and Random Forest) in combination with different types of descriptors (ISIDA fragments issued from Condensed Graphs of Reactions, MOLMAP, Electronic Effect Descriptors, and Chem. Development Kit computed descriptors) have been used. Models have good predictive performance in 3-fold cross-validation done three times: balanced accuracy varies from 0.7 to 1. Developed models are available for the users at http://infochim.u-strasbg.fr/webserv/VSEngine.html. Eventually, these were challenged to predict feasibility conditions for ∼50 novel Michael reactions from the eNovalys database (originally from patent literature).
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXnvFGnsQ%253D%253D&md5=205b5b0bd2af7f06f63174e2f6f799b1
    31. 31
      Lin, A. I.; Madzhidov, T. I.; Klimchuk, O.; Nugmanov, R. I.; Antipin, I. S.; Varnek, A. Automatized Assessment of Protective Group Reactivity: A Step toward Big Reaction Data Analysis. J. Chem. Inf. Model. 2016, 56 (11), 2140– 2148,  DOI: 10.1021/acs.jcim.6b00319
      31
      Automatized Assessment of Protective Group Reactivity: A Step Toward Big Reaction Data Analysis
      Lin, Arkadii I.; Madzhidov, Timur I.; Klimchuk, Olga; Nugmanov, Ramil I.; Antipin, Igor S.; Varnek, Alexandre
      Journal of Chemical Information and Modeling (2016), 56 (11), 2140-2148CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
      The authors report a new method to assess protective groups (PGs) reactivity as a function of reaction conditions (catalyst, solvent) using raw reaction data. It is based on an intuitive similarity principle for chem. reactions: similar reactions proceed under similar conditions. Tech., reaction similarity can be assessed using the Condensed Graph of Reaction (CGR) approach representing an ensemble of reactants and products as a single mol. graph, i.e., as a pseudomol. for which mol. descriptors or fingerprints can be calcd. CGR-based inhouse tools were used to process data for 142,111 catalytic hydrogenation reactions extd. from the Reaxys database. Results reveal some contradictions with famous Greene's Reactivity Charts based on manual expert anal. Models developed in this study show high accuracy (∼90%) for predicting optimal exptl. conditions of protective group deprotection.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhslemsL%252FM&md5=734545a7f2a3b34f49af6713d443b6be
    32. 32
      Segler, M. H.S.; Waller, M. P. Modelling Chemical Reasoning to Predict and Invent Reactions. Chem. - Eur. J. 2017, 23 (25), 6118– 6128,  DOI: 10.1002/chem.201604556
      32
      Modelling Chemical Reasoning to Predict and Invent Reactions
      Segler, Marwin H. S.; Waller, Mark P.
      Chemistry - A European Journal (2017), 23 (25), 6118-6128CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
      The ability to reason beyond established knowledge allows org. chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chem. reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph contg. 14.4 million mols. and 8.2 million binary reactions, which represents the bulk of all chem. reactions ever published in the scientific literature. Our model outperforms a rule-based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data-driven model generalises even beyond known reaction types, and is thus capable of effectively (re-)discovering novel transformations (even including transition metal-catalyzed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub-second time frame, the model can be used as a high-throughput generator of reaction hypotheses for reaction discovery.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2sXislersw%253D%253D&md5=ac3b304ec62b8d90110b7722305e2b3d
    33. 33
      Mikie Kanada, R.; Taniguchi, T.; Ogasawara, K. Asymmetric Hydrogen Transfer Protocol for a Synthesis of (+)-Frontalin and (−)-Malyngolide. Tetrahedron Lett. 2000, 41 (19), 3631– 3635,  DOI: 10.1016/S0040-4039(00)00430-5
      There is no corresponding record for this reference.
    34. 34
      Faroux-Corlay, B.; Clary, L.; Gadras, C.; Hammache, D.; Greiner, J.; Santaella, C.; Aubertin, A. M.; Vierling, P.; Fantini, J. Synthesis of Single- and Double-Chain Fluorocarbon and Hydrocarbon Galactosyl Amphiphiles and Their Anti-HIV-1 Activity. Carbohydr. Res. 2000, 327 (3), 223– 260,  DOI: 10.1016/S0008-6215(00)00055-0
      There is no corresponding record for this reference.
    35. 35
      Wang, H.; Yu, S. Synthesis of Isoquinolones Using Visible-Light-Promoted Denitrogenative Alkyne Insertion of 1,2,3-Benzotriazinones. Org. Lett. 2015, 17 (17), 4272– 4275,  DOI: 10.1021/acs.orglett.5b01960
      35
      Synthesis of Isoquinolones Using Visible-Light-Promoted Denitrogenative Alkyne Insertion of 1,2,3-Benzotriazinones
      Wang, Hao; Yu, Shouyun
      Organic Letters (2015), 17 (17), 4272-4275CODEN: ORLEF7; ISSN:1523-7052. (American Chemical Society)
      A visible-light-promoted regioselective denitrogenative insertion of terminal alkynes into 1,2,3-benzotriazinones is reported. This mechanistically novel process allows the synthesis of substituted isoquinolones in satisfactory isolated yields (24 examples, 46-84% yield) at room temp. under visible-light irradn. with the assistance of a photocatalyst. The proposed single-electron-transfer pathway was supported by TEMPO trapping, radical clock expts., and Stern-Volmer anal.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2MXhtlOjur3K&md5=86034a6db6d0456dbe780e5faec49e63
    36. 36
      Li, K.; Zeng, Y.; Neuenswander, B. Sequential Pd (II) -Pd (0) Catalysis for the Rapid Synthesis of Coumarins. J. Org. Chem. 2005, 70 (16), 6515– 6518,  DOI: 10.1021/jo050671l
      There is no corresponding record for this reference.
    37. 37
      Mavunkel, B.; Xu, Y.; Goyal, B.; Lim, D.; Lu, Q.; Chen, Z.; Wang, D.-X.; Higaki, J.; Chakraborty, I.; Liclican, A. Pyrimidine-Based Inhibitors of CaMKIIδ. Bioorg. Med. Chem. Lett. 2008, 18 (7), 2404– 2408,  DOI: 10.1016/j.bmcl.2008.02.056
      There is no corresponding record for this reference.
    38. 38
      Lautens, M.; Maddess, M. L. Chemoselective Cross Metathesis of Bishomoallylic Alcohols : Rapid Access to Fragment A of the Cryptophycins. Supplementary Material The Following Includes Representative Experimental Procedures and Details for Isolation of Compounds. Full Characterisat. Org. Lett. 2004, 6 (12), 1883– 1886,  DOI: 10.1021/ol049883f
      There is no corresponding record for this reference.
    39. 39
      Krüger, T.; Vorndran, K.; Linker, T. Regioselective Arene Functionalization: Simple Substitution of Carboxylate by Alkyl Groups. Chem. - Eur. J. 2009, 15 (44), 12082– 12091,  DOI: 10.1002/chem.200901774
      39
      Regioselective Arene Functionalization: Simple Substitution of Carboxylate by Alkyl Groups
      Krueger, Tobias; Vorndran, Katja; Linker, Torsten
      Chemistry - A European Journal (2009), 15 (44), 12082-12091CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
      Arenes with various alkyl side-chains were synthesized in high yields and excellent regioselectivities. The carboxylate group in toluic and naphthoic acids was conveniently substituted by alkyl halides by Birch redn. and subsequent decarbonylation. The method is characterized by inexpensive starting materials and reagents, and methylation of arenes was realized. Besides simple alkyl substituents, the scope of arene functionalization was extended by benzyl as well as fluoro-, amino-, and ester-contg. alkyl groups. The alkylation of 1-naphthoic acid during Birch redn. can be controlled by the addn. of tert-butanol. This allowed the regioselective synthesis of mono and bis-substituted naphthalenes from the same starting material.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXhtlyqu77F&md5=b143b98bc1a436b16bf909a9347629dc
    40. 40
      Palmes, J. A.; Paioti, P. H.S.; De Souza, L. P.; Aponick, A. PdII-Catalyzed Spiroketalization of Ketoallylic Diols. Chem. - Eur. J. 2013, 19 (35), 11613– 11621,  DOI: 10.1002/chem.201301723
      40
      PdII-Catalyzed Spiroketalization of Ketoallylic Diols
      Palmes, Jean A.; Paioti, Paulo H. S.; de Souza, Leonardo Perez; Aponick, Aaron
      Chemistry - A European Journal (2013), 19 (35), 11613-11621CODEN: CEUJED; ISSN:0947-6539. (Wiley-VCH Verlag GmbH & Co. KGaA)
      A high-yielding stereoselective method for forming spiroketals from simple ketone-allylic diol derivs. was reported. Using catalytic [PdCl2(MeCN)2] in THF at 0° these dehydration-cyclization reactions require only mild conditions to produce vinyl-substituted spiroketals in high yields after brief reaction times with water as the only byproduct. Using this method, the stereochem. information embedded at the nucleophile is transmitted down-the-chain and efficiently sets the stereochem. at both the anomeric carbon atom and the newly formed allylic stereocenter. The title compds. thus formed included a spiroketal (I) and related substances, such as ro[5.5]undecane derivs. and 1,6-dioxaspiro[4.5]decane derivs., carbohydrate monosaccharide analogs, such as a 5,9-anhydro-2,3,4-trideoxy-D-gluco-5-deculo-5,1-pyranose deriv. The synthesis of the target compds. was achieved by a palladium-catalyzed cyclization of a chiral 3,13-dihydroxy-11-tetradecen-7-one deriv. (II) and similar compds., for example (8E)-1,10-dihydroxy-8-decen-5-one.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3sXhtFShtLrN&md5=4271b9729abcbf0817cd65ea2cd5fb49
    41. 41
      Liu, J.; Fitzgerald, A. E.; Mani, N. S. Facile Assembly of Fused Benzo[4,5]Furo Heterocycles. J. Org. Chem. 2008, 73 (7), 2951– 2954,  DOI: 10.1021/jo8000595
      There is no corresponding record for this reference.
    42. 42
      Schaub, C.; Müller, B.; Schmidt, R. R. Sialyltransferase Inhibitors Based on CMP-Quinic Acid. Eur. J. Org. Chem. 2000, 2000 (9), 1745– 1758,  DOI: 10.1002/(SICI)1099-0690(200005)2000:9<1745::AID-EJOC1745>3.0.CO;2-8
      There is no corresponding record for this reference.
    43. 43
      Mikolov, T.; Sutskever, I.; Chen, K.; Corrado, G. S.; Dean, J. Distributed Representations of Words and Phrases and Their Compositionality. In Advances in neural information processing systems; NIPS: Lake Tahoe, 2013; pp 3111– 3119.
      There is no corresponding record for this reference.
    44. 44
      García-Alonso, C. R.; Pérez-Naranjo, L. M.; Fernández-Caballero, J. C. Multiobjective evolutionary algorithms to identify highly autocorrelated areas: the case of spatial distribution in financially compromised farms. Ann. Oper. Res. 2014, 219 (1), 187– 202,  DOI: 10.1007/s10479-011-0841-3
      There is no corresponding record for this reference.
    45. 45
      Open-source. RDKit: Open-Source Cheminformatics Software , 2006 (accessed on Sept 28, 2017).
      There is no corresponding record for this reference.
    46. 46
      Schneider, N.; Lowe, D. M.; Sayle, R. A.; Landrum, G. A. Development of a Novel Fingerprint for Chemical Reactions and Its Application to Large-Scale Reaction Classification and Similarity. J. Chem. Inf. Model. 2015, 55 (1), 39– 53,  DOI: 10.1021/ci5006614
      46
      Development of a Novel Fingerprint for Chemical Reactions and Its Application to Large-Scale Reaction Classification and Similarity
      Schneider, Nadine; Lowe, Daniel M.; Sayle, Roger A.; Landrum, Gregory A.
      Journal of Chemical Information and Modeling (2015), 55 (1), 39-53CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
      Fingerprint methods applied to mols. have proven to be useful for similarity detn. and as inputs to machine-learning models. Here, we present the development of a new fingerprint for chem. reactions and validate its usefulness in building machine-learning models and in similarity assessment. Our final fingerprint is constructed as the difference of the atom-pair fingerprints of products and reactants and includes agents via calcd. physicochem. properties. We validated the fingerprints on a large data set of reactions text-mined from granted United States patents from the last 40 years that have been classified using a substructure-based expert system. We applied machine learning to build a 50-class predictive model for reaction-type classification that correctly predicts 97% of the reactions in an external test set. Impressive accuracies were also obsd. when applying the classifier to reactions from an inhouse electronic lab. notebook. The performance of the novel fingerprint for assessing reaction similarity was evaluated by a cluster anal. that recovered 48 out of 50 of the reaction classes with a median F-score of 0.63 for the clusters. The data sets used for training and primary validation as well as all python scripts required to reproduce the anal. are provided in the Supporting Information.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXitFGrsrrE&md5=694e4122d2642de22065e46d7a0826cc
    47. 47
      Williams, R. J.; Zipser, D. A Learning Algorithm for Continually Running Fully Recurrent Neural Networks. Neural Comput. 1989, 1 (2), 270– 280,  DOI: 10.1162/neco.1989.1.2.270
      There is no corresponding record for this reference.
    48. 48
      Gobbi, A.; Poppinger, D. Genetic Optimization of Combinatorial Libraries. Biotechnol. Bioeng. 1998, 61 (1), 47– 54,  DOI: 10.1002/(SICI)1097-0290(199824)61:1<47::AID-BIT9>3.0.CO;2-Z
      48
      Genetic optimization of combinatorial libraries
      Gobbi, Alberto; Poppinger, Dieter
      Biotechnology and Bioengineering (1998), 61 (1), 47-53CODEN: BIBIAU; ISSN:0006-3592. (John Wiley & Sons, Inc.)
      Most agrochem. and pharmaceutical companies have set up high-throughput screening programs which require large nos. of compds. to screen. Combinatorial libraries provide an attractive way to deliver these compds. A single combinatorial library with four variable positions can yield >1012 potential compds., if one assumes that ∼1000 reagents are available for each position. This is far more than any high-throughput screening facility can afford to screen. We have proposed a method for iterative compd. selection from large databases, which identifies the most active compds. by examg. only a small fraction of the database. In this article, we describe the extension of this method to the problem of selecting compds. from large combinatorial libraries.
      >> More from SciFinder ®
      https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK1cXisVeksb8%253D&md5=3cfe243bf7cea6951886c5551f5ab3c6

  • Supporting Information

    Supporting Information

    The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscentsci.8b00357.

    • Description of computational methods, frequency vs rank plots, extended lists of reaction examples, and comparison of the model performance against baseline models (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.