Download Hi-Res ImageDownload to MS-PowerPointCite This:J. Chem. Inf. Model. 2012, 52, 11, 2864-2875

Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17

View Author Information

† Department of Chemistry and Biochemistry, NCCR TransCure, University of Berne, Freiestrasse 3, 3012 Berne, Switzerland

‡ Biomolecular Screening Facility, NCCR Chemical Biology, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

*Phone: +41 31 631 43 25. Fax: +41 31 631 80 57. E-mail: [email protected]

Cite this: J. Chem. Inf. Model. 2012, 52, 11, 2864–2875

Publication Date (Web):October 22, 2012

https://doi.org/10.1021/ci300415d

Article Views

37170

Altmetric

Citations

743

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.

PDF (3 MB)

Get e-Alerts

SUBJECTS:

Go to Journal of Chemical Information and Modeling

Get e-Alerts

Abstract

Drug molecules consist of a few tens of atoms connected by covalent bonds. How many such molecules are possible in total and what is their structure? This question is of pressing interest in medicinal chemistry to help solve the problems of drug potency, selectivity, and toxicity and reduce attrition rates by pointing to new molecular series. To better define the unknown chemical space, we have enumerated 166.4 billion molecules of up to 17 atoms of C, N, O, S, and halogens forming the chemical universe database GDB-17, covering a size range containing many drugs and typical for lead compounds. GDB-17 contains millions of isomers of known drugs, including analogs with high shape similarity to the parent drug. Compared to known molecules in PubChem, GDB-17 molecules are much richer in nonaromatic heterocycles, quaternary centers, and stereoisomers, densely populate the third dimension in shape space, and represent many more scaffold types.

Introduction

ARTICLE SECTIONS

Jump To

The cumulated efforts of synthetic chemistry over the last century has produced over 60 million compounds as collected by Chemical Abstracts Service. (1, 2) Since the implementation of combinatorial and parallel synthesis by academic and industrial drug discovery, the number of druglike small molecules (organic compounds of intermediate polarity with MW ≤ 500 Da) has increased even further. (3-5) The combined corporate, academic, and commercial collections worldwide probably total over 100 million different small molecules. (6) Despite these impressive numbers, it has become increasingly difficult to develop new small molecule drugs, largely due to lack of efficacy, side effects, and toxicity issues. (7, 8)De novo drug design (9-12) may help to address this problem by investigating even much larger numbers of yet unknown molecules by virtual screening (13-15) in search of innovative structures that might exhibit improved selectivity and ADMET profiles.

The majority of de novo drug design methods generate molecules within genetic algorithms that optimize a desired property such as a docking score by evolving a molecule population through breeding and mutation cycles. In most cases these algorithms generate new molecules by recombining known building blocks with known reactions, which severely limits their innovative potential. To circumvent this limitation, we recently approached the direct enumeration of chemical space by extending an approach to de novo design pioneered by Cayley, the inventor of graph theory, to count acyclic hydrocarbons (16) and later used in computer assisted structure elucidation. (17-19) The idea is to enumerate molecules from first principles starting from mathematical graphs irrespective of pre-existing building blocks to avoid a historical bias in structure selection. Geometrical strain and functional group stability criteria are used to ensure that the molecules produced are chemically meaningful. By this method we obtained the chemical universe database GDB-11 enumerating 26.4 million different molecules up to 11 atoms of C, N, O, and F (110.9 million molecules when including stereoisomers). (20, 21) The number increased to almost 1 billion (not counting stereoisomers) for GDB-13 listing all molecules up to 13 atoms of C, N, O, Cl, and S. (22, 23) Both databases were later shown to be useful sources of molecular diversity to discover new receptor ligands by virtual screening, synthesis, and testing. (24-30)

While GDB-11 and GDB-13 uncovered impressive numbers of possible molecules, the databases only addressed very small organic molecules (MW < 200 Da), which are of interest as relatively small fragments (31) but rarely correspond to actual drugs. Herein we report the enumeration of organic molecules up to 17 atoms of C, N, O, S, and halogens, forming the chemical universe database GDB-17 containing 166.4 billion organic molecules. GDB-17 reaches into molecular sizes compatible with many drugs (367 approved drugs ≤17 atoms) and typical for lead compounds (100 < MW < 350 Da). (32) Millions of isomers of known drugs are readily identified in GDB-17. While molecules up to 17 atoms in the public databases PubChem, (33) ChEMBL, (34) or DrugBrank (35) are mostly achiral, aromatic, and heteroaromatic compounds with rodlike shapes, GDB-17 molecules are mostly nonaromatic heterocycles with many quaternary centers and stereoisomers. GDB-17 densely populates the third dimension in shape space and represents many more scaffold types than found in PubChem.

Results and Discussion

ARTICLE SECTIONS

Jump To

Enumeration

The enumeration followed the approach used for GDB-13 starting from the complete list of graphs as given by the program GENG. (36) Graphs corresponding to unstrained hydrocarbons were selected based on geometrical criteria and expanded to ″skeletons″ (unsaturated hydrocarbons) by substituting bonds (single, double, triple bonds) for graph edges. These skeletons were themselves expanded to molecules by substituting atoms (C, N, O, etc.) for graph nodes, respecting valency rules, and eliminating chemically unstable and problematic functional groups (Figure 1). The code used for GDB-13, which stalled at 14 or more atoms due to inefficient memory usage, was entirely rewritten considering process design and efficiency. New graph and molecule selection criteria were also added to constrain the combinatorial explosion above 13 atoms. The code redesign resulted in a 400-fold increase in computing speed allowing to complete the enumeration up to 17 atoms of C, N, O, S, and halogens within reasonable time, as detailed below.

GENG (36) was set to enumerate all graphs up to 17 nodes with maximum valency of four (quaternary carbon) considering only topologically planar (i.e., eliminating knotted topologies corresponding to the K₅ and K_3,3 graphs), connected (e.g., no catenanes) graphs, which returned 114,304,569,097 graphs. The graphs were then converted to hydrocarbons substituting carbon atoms for graph nodes and carbon–carbon single bonds for graph edges. Hydrocarbons were selected for limited ring strain and topological complexity by applying the hydrocarbon filters H1 to H5 (Table 1), which left 5,422,153 hydrocarbons (0.005% of the graphs) to be considered for molecule generation. These 5.4 million hydrocarbons were converted to ″skeletons″ by introducing double and triple bonds following valency rules and the unsaturation filters S1 to S5, which primarily restrict ring strain and reactive unsaturations (Table 2). The selected skeletons were then run through an aromatization–dearomatization cycle, and duplicates were removed, which eliminated aromatic tautomers. The introduction of unsaturations generated on average 246 skeletons per graphs for a total of 1,330,958,530 skeletons.

The 1.3 billion skeletons were diversified into molecules by combinatorially substituting nitrogen and oxygen for carbon following valency rules but not generating any heteroatom–heteroatom bond. All generated molecules were then checked for undesirable functional groups (FG) to ensure that the molecules have a good probability to be stable and synthetically accessible (FG filters F1–F12, Table 3). These filters followed in part previously reported criteria for removing problematic functional groups from screening libraries. (37, 38) This diversification of skeletons into ″CNO″ molecules produced 110.4 billion molecules, corresponding to an average of 83 molecules per skeleton and 20,400 molecules per graph. Postprocessing steps P1–P7 were finally implemented for additional diversity by combinatorial atom type substitutions (Table 4). These postprocessing steps added another 56 billion molecules, resulting in a total of 166.4 billion molecules in the complete GDB-17. The overall molecule generation procedure was always completed in one run for each graph, and all molecules generated from each graph were checked for duplicates to guarantee that all GDB-17 molecules are different (Table 5). The overall computation consumed 100,000 CPU hours, which is only 2.5-fold more than the computing time originally invested for GDB-13.

Table 1. Ring Strain and Complexity Filters To Select Hydrocarbon Graphs

filter	description	comment
H1	SAV (″smallest atomic volume″): all graphs ≤ C₁₁ are converted to a 3D-structure, and the volume of the tetrahedron around each C atom is checked for a minimum value.	95.2% of graphs ≤ C₁₁ are discarded due to failed 3D-conversion (using CORINA or ChemAxon molconverter) or due to distorted (planar, pyramidal) centers. Simple polyhedra such as cubane are preserved. See the SI of ref 22 for details.
H2	NA2SR (″no atom shared by two small rings″): removes graphs ≥ C₁₂ containing fused or spiro linkages between 3- or 4-membered rings.	The vast majority of fused small ring systems are highly strained and reactive. 96.7% of C₁₂- and 97.3% of C₁₃-graphs are removed. See also filter H3.
H3	NBH3R (″no bridgehead in 3 rings″): Graphs ≥ C₁₄ with three or more ″nonzero″ bridgehead atoms shared by three or more rings are removed.	These multilooped topologies would correspond to molecules of high synthetic complexity. 99.60% of C₁₄-, 99.77% of C₁₅-, 99.95% of C₁₆-, and 99.95% of C₁₇-graphs are removed by filters H2+H3.
H4	1SR (″one small ring″): C₁₅ and C₁₆ graphs are allowed at most one 3- or 4-membered ring.	71.89% of the C₁₅- and 79.50% the C₁₆-graphs that passed H2+H3 are removed.
H5	0SR (″no small ring″): C₁₇-graphs are not allowed any small rings.	96.95% C₁₇-graphs that passed H2+H3 are removed.

Table 2. Unsaturation Filters To Enumerate Skeletons (Unsaturated Hydrocarbons)

filter	description	comment
S1	no allenes (C═C═C)	Although known and sometime found in bioactive molecules, allenes are usually reactive and quite difficult to prepare but combinatorially extremely frequent.
S2	no unsaturations in 3-membered rings	Cyclopropenes are known but quite reactive and difficult to prepare. Cyclopropynes are unstable.
S3	at most one sp²-center in 4-membered rings	Cyclobutenes and cyclobutynes are not enumerated, but the skeletons leading to β-lactams and β-lactones are generated.
S4	triple bonds restrictions	No triple bond in 3- or 4-membered rings, max. One triple bond in rings ≥9 and max two triple bonds in ≥11 rings. Only terminal triple bonds for C₁₇-hydrocarbons (allowing to generate nitriles).
S5	bridgehead double bond restrictions	If a “non-zero” bridgehead carbon is sp², the ring sizes of the smallest set of smallest rings will be checked. At least one ring of this carbon must be ≥8. In case of two such bridgeheads are sp², the ring size must be ≥10.

Table 3. Functional Group Filters To Enumerate CNO Moleculesa

filter	description	comment
F1	XCX: only one N or O next to a sp³ carbon or two oxygens if both oxygens are ring atoms.	Aminals, hemiacetals gem-diols, and acyclic acetals are not enumerated. Only cyclic acetals are allowed.
F2	Maximum one N or O in small rings.	Allows epoxides, oxiranes, aziridines, azetidines but no cyclic acetals inside 4-membered rings.
F3	Anhydrides (O═C)–O–(C═O) are removed.	Most anhydrides are unstable toward hydrolysis.
F4	Acetal chains O–Csp³–O–Csp³–O are removed.	Although sometimes found, acetal chains are diffult to plan synthetically.
F5	Molecules with a primary amine and a ketone or aldehyde are removed.	This combination often polymerizes.
F6	C═N are removed unless the sp² carbon is connected to a further N or O atom.	Removes imines which are unstable but retains amidines and guanidines.
F7	(N/O)–C═N–C═(N/O) are removed.	The corresponding (N/O)═C–N–C═(N/O) tautomer is allowed.
F8	Enol/enamine: removes O or N atoms adjacent to a nonaromatic C═C.	Enols, enamines, enol ethers, etc. are almost always unstable toward hydrolysis to the parent carbonyl compound.
F9	Acyclic carbonates C–O–(C═O)–O–C are removed.	Acyclic carbonates are rather unstable toward hydrolysis.
F10	Carbonic acids (O–CO₂H), carbamic acids (N–CO₂H), and β-carboxylic acid ((C═O)–C–CO₂H) are removed.	These FG decarboxylate spontaneously.
F11	Bridgehead amides: If a “non-zero” bridgehead nitrogen is bound to a nonaromatic sp² atom, the ring sizes of the smallest set of smallest rings will be checked. At least one ring of the nitrogen must be ≥9.	Such amides are ″twisted″ and nonconjugated and therefore quite unstable toward hydrolysis.
F12	C═C: Molecules of 17 atoms with nonaromatic carbon–carbon unsaturations are removed.	Nonaromatic C═C are highly frequent but often reactive toward polymerization, cycloadditions, isomerizations, oxidation, or nucleophilic addition.

No heteroatom–heteroatom bonds are generated at all.

Table 4. Postprocessing Steps for Aromatic Heterocycles, Oximes, Nitro, CF₃, Halogens, and Sulfur

step	description	comment
P1	Aromatic C to N: aromatic C atoms adjacent to an aromatic N or O atom are converted to N if valency allows.	Aromatic heterocycles with heteroatom–heteroatom bonds are created e.g. 1,2-oxazoles from furans, 1,2,3-triazoles from imidazoles.
P2a	Ketone oximes C═N–OH: ketones are converted to oximes.	Note that alkylated oximes, hydroxamates, hydrazides, and hydrazone are not considered.
P3	Aromatic halogens: aromatic OH groups are changed to halogens.	Halogen = F, Cl, Br, I, max. Two Br or I per aromatic ring.
P4	Trifluoromethyls: tert-butyl groups are changed to CF₃.
P5	Aromatic nitro groups: aromatic CO₂H are converted to NO₂.	Aliphatic nitro groups are not considered.
P6	Thiophenes: sulfur was substituted for all heteroaromatic oxygen atoms.	Aliphatic thiols and thioethers are not considered.
P7a	Sulfones: carbonyl groups (C═O) in ketones, acids, carboxamides, and carbamates are changed to SO₂.	Note that C═S and sulfoxides (S═O) are not generated.

Steps P2 and P7 increase the heavy atom count (hac), generating for example some 17 atoms molecules with small rings. All molecules with hac >17 were removed to avoid a combinatorial explosion.

Table 5. Database Generation Statistics

HAC	filtersa	graphsb	hydrocarbonsc	skeletonsd	moleculese	CPU, hf
1	SAV, FG	1	1	1	3	0
2		1	1	3	6	0
3		2	2	4	14	0
4		6	4	12	47	0
5		20	10	32	219	0
6		74	31	119	1,091	0
7		321	98	448	6,029	0
8		1,663	370	2,004	37,435	0
9		9,616	1,448	9,472	243,233	0
10		61,840	6,325	48,721	1,670,163	0
11		427,135	29,496	264,321	12,219,460	3
12	NA2SR	3,120,002	104,165	1,188,127	72,051,665	18
13		23,722,244	651,850	7,370,864	836,687,200	206
14	NBH3R	186,092,397	752,277	27,419,837	2,921,398,415	856
15	1SR	1,496,007,875	960,415	118,977,963	15,084,103,347	5,378
16		12,176,341,897	1,331,875	213,259,331	38,033,661,355	14,415
17	0SR, C═C	100,418,784,003	1,583,786	962,417,271	109,481,780,580	79,259
SUM		114,304,569,097	5,422,154	1,330,958,530	166,443,860,262	100,134

See Tables 1–4 for details.

Graphs produced by GENG for planar, connected graphs up to 17 nodes with maximum node valence of four.

Hydrocarbons generated from graphs and passing the filters in Table 1 for limited ring strain and complexity.

Unsaturated hydrocarbons generated from hydrocarbons using filters in Table 2.

Molecules generated from hydrocarbons by adding heteroatoms (Table 3 and 4), as 2D-structures and stored as SMILES.

Computation was parallelized on 360 CPU.

Comparing GDB-17 with PubChem, ChEMBL, and DrugBank

One of the key questions arising from the systematically enumerated chemical space available in GDB-17 is whether this collection significantly differs from the already known chemical space. To perform this comparison, we collected molecules up to 17 atoms in the public archives PubChem, (33) ChEMBL, (34) and DrugBank, (35) to form the reference collections up to 17 atoms PubChem-17 (2,526,453 cpds), ChEMBL-17 (89,156 cpds), and DrugBank-17 (367 cpds, approved drugs only). ChEMBL-17 and DrugBank-17 represent subsets of the larger PubChem-17 which are focused on molecules with biological activities that are reported (ChEMBL-17) respectively clinically approved (DrugBank-17). Unfortunately commercial collections such as the CAS or Beilstein archives could not be obtained. However considering that the 2.5 million molecules in PubChem-17 represent approximately 10% of the 25 million unique molecules listed in PubChem, CAS-17 probably contains around 10% of the entire CAS archive, (1, 2) i.e. slightly more than 6 million molecules (2.4-fold larger than PubChem-17).

The comparison of the enumerated chemical space with known molecules starts with considering database size. Thus, GDB-17 (166.4 billion molecules) is much larger than the sum of all known molecules of similar size as found in PubChem-17 (0.001% of GDB-17). The size of GDB-17 originates in the increase of possible molecules as a function of heavy atom count, which is exponential and much steeper than in the reference databases (Figure 2A). As a consequence the MW range of GDB-17 shows a sharp peak at 240 < MW < 250 Da. The same distribution is observed in the leadlike subset GDBLL-17 (29 billion structures, see below) and leadlike/no small ring subset GDBLLnoSR-17 (22 billion structures, see below), while the MW distribution in the reference databases is more even (Figure 2B).

GDB-17 contains an impressive number of molecules in the area of known drugs. For example, millions of isomers can be identified in GDB-17 for fifteen typical marketed drugs of 14 to 17 atoms selected from DrugBank-17 (Table 6, Figure 3). The examples shown in Figure 3 were selected among isomers with a high shape similarity to the parent drug as measured by the OpenEye scoring function ROCS (Rapid Overlay of Chemical Structures), a well validated virtual screening tool to identify bioactive analogs. (39, 40) These isomers include obvious variations of the parent structure such as ″methyl walk″ analogs, for example structures 5 and 11 as isomers of drugs 4 and 10, as well as nontrivial changes such as different aromatic heterocycles (2, 3, 8, 9, 17, 18, 35, 38, 39, 46), different ring size and connectivity (15, 23, 24, 26, 27, 32, 33, 39, 41, 44–48), and different functional groups (14, 29, 30, 32, 33, 36, 39, 42, 48).

On the other hand, GDB-17 represents a selective enumeration and therefore does not contain all molecules in the reference databases. Overall 57% of PubChem-17, 60% of ChEMBL-17, and 68% of DrugBank-17 are compatible with the GDB-17 enumeration rules. The molecules found in the reference databases but not considered for GDB-17 contain nonenumerated features such as certain types of halogens (e.g., aliphatic halogens) or sulfurs (thiols, thioethers, thioureas), functional groups (e.g., acyclic acetals, hemiacetals, aminals, azides, aliphatic nitro groups), elements (P, Si, B, etc.), skeletons (nonaromatic C═C), or graphs (e.g., spiro-fused cyclopropanes) (Figure 4A).

Figure 2. Size and MW profiles of the enumerated chemical space in GDB and the reference databases PubChem, ChEMBL, and DrugBank. The size of the leadlike subsets of GDB (GDBLL, GDBLLnoSR) is extrapolated from analyzing a 1% random subset of GDB-17.

Table 6. Drug Isomers Found in GDB-17

drug namea	elemental formula	no. of isomer
Acyclovir	C₈H₁₁N₅O₃	8,132,952
Aminoglutethimide	C₁₃H₁₆N₂O₂	183,901,628
Aminophenazone	C₁₃H₁₇N₃O	97,853,936
Dexmedetomidine	C₁₃H₁₆N₂	9,721,191
Diethylcarbamazine	C₁₀H₂₁N₃O	22,409
Ethoxzolamide	C₉H₁₀N₂S₂O₃	4,563,491
Felbamate	C₁₁H₁₄N₂O₄	369,751,288
Fencamfamine	C₁₅H₂₁N	53,917,207
Guanadrel	C₁₀H₁₉N₃O₂	60,319,220
Procaine	C₁₃H₂₀N₂O₂	476,975,898
Sulfadiazine	C₁₀H₁₀N₄SO₂	17,003,297
Tinidazole	C₈H₁₃N₃SO₄	24,575,941
Tizanidine	C₉H₈N₅SCl	109,635
Trioxsalen	C₁₄H₁₂O₃	1,800,849
Varenicline	C₁₃H₁₃N₃	19,676,640

See Figure 3 for structural formula of the drugs and examples of isomers.

Figure 3. Drugs and examples of isomers found in GDB-17. All isomers shown have a shape similarity score ROCS > 1.4. None of the isomers shown are known (Scifinder search). Only acyclovir does not occur in GDB-17 because it contains a hemiaminal (N–Csp³–O), a functional group which is excluded from the enumeration.

Figure 4. Molecule topologies and categories in GDB-17 and reference databases. A. Percentage of reference database compatible with GDB-17 enumeration rules or excluded due to nonenumerated halogen (acyl halide, aliphatic halocarbons) or sulfur (thiols, thioethers), functional groups (acyclic acetals, hemiacetals, aminals, azides, aliphatic nitro groups), element (P, Si, B, Bi, Hg, etc.), skeleton (nonaromatic C═C), or graph (e.g., small rings at 17 atoms). B. Fraction of compounds with small rings. C. Topologies D. Database contents as function of molecular categories. Molecules are assigned to one category only with priority order heteroaromatic > aromatic > heterocyclic > carbocyclic > acyclic. The data for GDB-17 and its subsets were computed from a 1% random subset of the database.

Small Rings, Topology, and Compound Categories

The most striking difference between the enumerated chemical space in GDB-17 and known molecules resides in the occurrence of small rings (3- or 4-membered ring). Small rings are very frequent in the systematic enumeration of graphs and the resulting molecules. However they are also relatively difficult to synthesize and often unstable, and indeed they are not found very often in known molecules. While only 4–6% of the compounds in the reference databases contain small rings, the enumerated chemical space up to 16 atoms is to 83% a small ring compound database (Figure 4B). The fraction of small ring compounds in GDB falls to 8.2% at 17 atoms because no small ring were allowed in 17 node graphs (small ring molecules at 17 atoms stem from atom-adding postprocessing steps such as the transformation of carbonyls to sulfonyls and of ketones to oximes, Table 5). Nevertheless the majority (66%) of GDB-17 are molecules with 17 atoms, and the low percentage of small ring compounds at 17 atoms results in an overall 28% of small ring compounds in GDB-17 (25% in GDBLL-17).

In terms of topology, all databases contain approximately two-thirds of molecules with two or three cycles (Figure 4C). Key differences occur in acyclic compounds, which are relatively rare in GDB-17 (1.8%, 3.0 billion molecules) but make up 25% of DrugBank-17. Tri- and polycyclic molecules furthermore combine to 32% of GDB-17 but are much less frequent in the reference databases (PubChem-17: 7%; ChEMBL-17: 16%, DrugBank-17: 6%). The leadlike subset is also rich in tri- and polycyclic compounds (GDBLL-17: 33%), but their proportion is reduced when small rings are removed (GDBLLnoSR-17: 22%).

In terms of compound categories, heteroaromatic compounds make up a large third of GDB-17 and the reference databases (Figure 4D). By contrast aromatics, which also make up a third of reference databases, are quite rare in GDB-17 (0.8%, 1.3 billion molecules). GDB-17 is instead much richer in nonaromatic heterocycles (GDB-17: 57%, GDBLL-17: 41%, GDBLLnoSR-17: 35%) than the reference databases of known compounds (PubChem-17: 12%, ChEMBL-17: 10%, DrugBank-17: 12%).

Polarity and Leadlikeness

The histograms of the calculated octanol:water partition coefficient clogP shows that GDB-17 and DrugBank-17 contain more polar molecules than PubChem-17 and ChEMBL-17 (Figure 5A/B). A similar effect is visible in other polarity descriptors such as the number of H-bond donor atoms (Figure 5C/D). The fact that polar molecules often require longer syntheses and are more difficult to purify than apolar ones might explain their lower proportion in PubChem and ChEMBL, which contain mostly synthesized molecules, compared to GDB-17 representing the spectrum of possibilities. The frequency of polar molecules in the systematic enumeration of GDB-17 results in only 18% of GDB-17 being leadlike compounds as defined by the value ranges 1 < clogP < 3 and 100 < MW < 350 Da. (32) These 18% correspond to 29 billion molecules defining the GDBLL-17 subset, 22 billion of which do not have small rings and form the GDBLLnoSR-17 subset. By comparison approximately half of the compounds from the reference databases are leadlike (PubChem-17: 47%; ChEMBL-17: 49%; DrugBank-17: 36%).

Figure 5. Polarity features. A. c logP histogram in intervals −5.5 to −4.5, −4.5 to −3.5, etc; B. Average clogP as function of hac; C. H-bond donor atom (HBD) histogram; D. Average HBD as function of hac. The data for GDB-17 and its subsets were computed from a 1% random subset of the database.

Molecular Shape

Organic molecules can be classified in terms of shape by analyzing the principal moments of inertia of their 3D structure, which allows to classify molecules either as rods (linear shape, e.g. stretched alkanes), discs (cyclic planar shape, e.g. benzene), or spheres (globular shape, e.g. cubane or adamantane). This analysis shows that the vast majority of currently used druglike molecules are either rodlike or disklike. Only a minority of the molecules used in medicinal chemistry possess any significant third dimension, leading to shape considerations as a design criteria for screening libraries. (41) Closer analyses of successes and failures show that molecules with a significant third dimension in shape are indeed often more successful in drug development programs, suggesting an “escape out of flatland” as a valuable strategy to search for better drug molecules. (42, 43) Nonplanarity is also more pronounced in natural products (NP) and products from diversity-oriented synthesis (DC) compared to commercial screening compounds (CC) and probably contributes to the higher protein binding selectivity of NP and DC compared to CC as observed in small molecule microarray experiments. (44, 45)

A 16.7 million random subset of GDB-17 was subjected to the above shape analysis, and the results were compared with the data for the reference databases. The analysis showed that GDB-17 molecules significantly populate the third dimension, which implies that the ″escape out of flatland″ is statistically unavoidable when considering the enumerated chemical space (Figure 6). The shape distribution into the third dimension is similar for the 29 billion leadlike subset GDBLL-17 and the 22 billion leadlike subset without small rings GDBLLnoSR-17. By comparison the known molecules in PubChem-17, ChEMBL-17, and DrugBank-17 are essentially rods and discs with relatively few spherical molecules. This “flatness” is a direct consequence of the abundance of aromatic systems in these databases of known compounds. Conversely, the occurrence of 3D-shaped molecules in GDB-17 results from the low proportion of acyclic and aromatic compounds and the high frequency of saturated heterocycles in the enumerated chemical space. GDB-17 molecules also contain more quaternary carbon centers (qv, Figure 7A/B) and bonds in fused rings (bfr, Figure 7C/D) compared to known compounds, which are features strongly associated with nonplanarity. The decrease in bfr at 14, 15, and 17 atoms in GDB reflects the introduction of the “no bridgehead in 3-rings”, “one small ring”, and “no small ring” filters which strongly reduce the number of topologies with high bfr. Somewhat unexpectedly, the small-ring filters also reduce the number of quaternary centers per molecule, as seen by the fact that the GDBLLnoSR subset contains fewer quaternary centers than GDB and its leadlike subset.

Figure 6. Molecular shape analyzed by the principal moments of inertia. (41) Occupancy maps are shown in the (P1,P2)-plane, in which P1 and P2 are the normalized ratios of the principal moments of inertia (for details see section Methods), and are colored from blue (1 cpd/pixel) to purple (maximum cpd/pixel for each map: GDB-17: 4,691, GDBLL-17: 889, GDBLLnoSR-17: 684, Pubchem-17: 6202, Chembl-17: 487, Drugbank-17: 4). The inserts show an enlarged view of the lower left edge of each triangle where occupancy is highest for PubChem-17, ChEMBL-17, and DrugBank-17. The GDB-17, GDBLL-17, and GDBLLnoSR-17 were analyzed with a random subset of 16.7 million molecules from GDB-17. For all compounds a single stereoisomer was analyzed as generated by CORINA.

Figure 7. Histograms of quaternary centers (qv) and bonds in fused rings (bfr) in the different databases. The data for GDB-17 and its subsets were computed from a 1% random subset of the database.

Stereochemistry

The much higher frequency of nonplanar molecules in GDB-17 compared to the reference databases should also be reflected in a larger number of stereocenters and hence possible stereoisomers per compound. The number of possible stereoisomers per molecule was determined using the 3D-generator CORINA, (46) which exhaustively generates stereoisomers from 2D structures. CORINA correctly excludes impossible combinations of stereoisomer flipping (e.g., only one stereoisomer for norbornane). CORINA produces enantiomers as pairs with the exception of atropisomers, which are rather rare, implying that molecules for which only a single diastereoisomer is produced are almost always achiral. Their number provides a lower estimate of the number of achiral molecules because meso compounds (e.g., 1,4-dimethylcyclohexane or (R,S)-2,3-butanediol) and achiral Z/E isomer pairs are not singled out.

The stereoisomer counting with CORINA was performed on the 16.7 million subset of GDB-17 and on the reference databases (Figure 8). GDB-17 molecules produced an average of 6.4 stereoisomers per molecule (GDBLL-17: 5.7 stereoisomers/cpd, GDBLLnoSR: 5.1 stereoisomers/cpd), which is three times more than in the reference databases (PubChem-17: 2.0 stereoisomers/cpd, ChEMBL-17: 2.0 stereoisomers/cpd, DrugBank-17: 2.1 stereoisomers/cpd). More than half of the molecules in the reference databases have only one stereoisomer (PubChem-17: 56%, ChEMBL-17: 58%, DrugBank-17: 55%), while only 5% are molecules with eight or more possible stereoisomers (PubChem-17: 4.1%, ChEMBL-17: 4.6%, DrugBank-17: 5.2%). By contrast GDB-17 (respectively GDBLL-17, GDBLLnoSR-17) contains only 22% (respectively 23%, 27%) of molecules with a single stereoisomer but 44% (respectively 38%, 32%) of molecules with eight of more stereoisomers. The smaller average number of stereoisomers per compound as a function of hac in GDBLLnoSR-17 compared to GDBLL-17 shows that the presence of small rings is partly responsible for the larger number of stereoisomers in GDB-17 compared to known compounds.

Figure 8. Stereochemistry. A Numbers of stereoisomers per compounds. B. Average number of stereoisomer per compound as a function of hac. Stereoisomers were generated from SMILES using CORINA. The data for GDB-17, GDBLL-17, and GDBLLnoSR-17 stem from the analysis of a random 16.7 million subset of GDB-17.

Novelty and Scaffolds

The above analyses show that the novelty of GDB-17 compared to PubChem-17 can be assigned in part to global structural features including the relative rarity of aromatic and acyclic compounds, the frequent occurrence of molecules with small rings and nonaromatic heterocycles, and the higher proportion of polar molecules (clogP < 0). GDB-17 molecules also differ from PubChem-17 molecules in that they contain more structural features leading to 3D-shapes, such as quaternary centers and bonds in fused rings, as well as generally more stereoisomers per molecule. Nevertheless GDB-17 contains impressive numbers of compounds within any constraints, as exemplified with the millions of isomers of known drugs and the size of the leadlike/no small rings subset GDBLLnoSR-17 containing 22 billion molecules. By their number these molecules are necessarily new although they represent variations of known compound types.

One can also analyze the databases for novelty independent of global parameters by focusing on the occurrence of “scaffolds”. As “scaffolds” we considered either the “Murcko scaffolds”, (47) which are defined as the saturated hydrocarbon graph of a molecule pruned of any terminal atom, or “ring systems” defined as hydrocarbon graphs without acyclic bonds. (21) The analysis was performed for GDB-17 by considering the 5.4 million unique graphs used for molecule generation (Table 1) and extracting graphs corresponding to Murcko scaffolds and ring systems. For PubChem-17 each molecule was converted to its parent saturated hydrocarbon. All terminal atoms were then removed iteratively to produce “Murcko scaffolds”, and all acyclic bonds were removed to produce “ring systems”. Each resulting list was reduced to unique structures by removing duplicates. Each series was split into three categories by analyzing the smallest set of smallest rings as follows: a) scaffolds containing at least one small ring (″SR″); b) scaffolds containing only 5–7 membered rings (″5–7″); and c) scaffolds without small rings containing at least one 8-membered or larger ring (″8+″). The scaffolds were further subdivided according to the number of quaternary centers (Table 7).

The scaffold and ring system analysis shows that GDB-17 contains 35-fold more Murcko scaffolds and 61-fold more ring systems than PubChem-17. The majority of the imbalance stems from scaffolds containing small rings (Murcko scaffolds: 52-fold excess in GDB-17, ring systems: 109-fold excess in GDB-17), in particular small ring scaffolds with quaternary centers (Murcko scaffolds: 105-fold excess in GDB-17, ring systems: 170-fold excess in GDB-17). If considering only Murcko scaffolds or ring systems with 5- to 7-membered rings and without any quaternary center, which are the easiest to synthesize and most common ring systems, the number of scaffolds is only, but still, 2-fold larger in GDB-17 compared to PubChem-17. Ring systems that are yet unknown even as substructure are readily identified in GDB-17, such as the yet unknown C₁₇-hydrocarbon graphs 49–55 shown in Figure 9.

Table 7. Scaffold Analysis of GDB-17 and PubChem-17a

Murcko scaffolds
no. of quat. C	0	1	2	>2	SUM
GDB-17, SR	19,804	56,975	65,536	42,895	185,210
GDB-17, 5–7	1,736	2,561	1,195	217	5,709
GDB-17, 8+	1,113	466	44	0	1,623
SUM	22,653	60,002	66,775	43,112	192,542
PubChem-17, SR	1,997	1,114	405	56	3,572
PubChem-17, 5–7	960	562	121	3	1,646
PubChem-17, 8+	307	41	8	0	356
SUM	3,264	1,717	534	59	5,574

Ring Systems
no. of quat. C	0	1	2	>2	SUM
GDB-17, SR	12,607	45,419	60,720	42,293	161,039
GDB-17, 5–7	1,135	2,143	1,126	217	4,621
GDB-17, 8+	978	426	44	0	1,448
SUM	14,720	47,988	61,890	42,510	167,108
PubChem-17, SR	600	521	314	45	1,480
PubChem-17, 5–7	480	375	111	3	969
PubChem-17, 8+	254	36	8	0	298
SUM	1,334	932	433	48	2,747

Murcko scaffolds are hydrocarbon graphs without any terminal atoms and ring systems are hydrocarbon graphs without any acyclic bonds. Scaffolds and ring systems are divided into three categories: SR: at least one small (3- or 4-membered) ring; 5–7: containing only 5- to 7-membered rings; 8+: no small ring and at least one 8-membered or larger ring. Rings are analyzed in the smallest set of smallest rings i.e.. bicyclo[2.2.1]heptane (norbornane) contains two 5-membered rings, while its 6-membered ring is not considered.

Figure 9. Examples of yet unknown C₁₇-ring systems from GDB-17. These hydrocarbons do not give any hits in Scifinder using ″any atom″ types for carbons and ″any bond″ for bonds, including substructure searches but locking further ring fusions. Stereochemistry is not considered in these searches. The ring systems are shown as one possible stereoisomer.

Conclusion

ARTICLE SECTIONS

Jump To

In summary the enumeration of organic molecules starting from mathematical graphs was realized up to 17 atoms of C, N, O, S, and halogens, yielding 166.4 billion molecules corresponding to a defined set of functional group and atom types. Compared to the 2.5 million known molecules up to 17 atoms found in PubChem, GDB-17 molecules contain generally more rings, in particular small rings, as well as many nonaromatic heterocycles. On the other hand, cyclic and aromatic compounds form a much smaller fraction of the database compared to PubChem. GDB-17 molecules furthermore contain more quaternary centers and bonds in fused rings than PubChem molecules, resulting in significant 3D-shapes and a larger number of stereoisomers per molecule. GDB-17 molecules are on average also more polar than known molecules (clogP < 0), although a leadlike subset occupying the range 0 < clogP < 3 still contains 22 billion molecules even when excluding small ring compounds. The structural diversity of GDB-17 is evidenced by the presence of a much larger number of scaffolds compared to known molecules. The abundance of nonplanar molecules suggests that the enumerated chemical space might serve as a rich source of inspiration to design new molecular series for drug discovery.

As to the size of GDB-17, working with 166.4 billion structures is challenging and currently not applicable to advanced virtual screening methods such as shape-based analyses or docking, which are computationally relatively intensive. For such applications a randomly selected subset of GDB-17 of a few hundred thousand to a few million structures is statistically significant and can be used as representative of the whole database. On the other hand, the identification of single molecules such as the selected analogs of known drugs shown in Figure 3, or the examples of polycyclic hydrocarbons shown in Figure 9, can deliver many more interesting results with the complete database because every single molecule is different and identifiable in its own right. The assembly of a searchable version of the entire GDB-17 database and its use for identifying drug analogs by virtual screening represented a challenge of its own and will be described in a separate publication.

Methods

ARTICLE SECTIONS

Jump To

General

All code packages were written in Java 1.6 with Jchem Libaries from ChemAxon. Every filter was applied to the imported molecule to define bond and atom positions of functional groups. All computations were parallelized on a 360-CPU cluster and manually controlled (100,000 CPU hours corresponds to 11 CPU years). Every step was completed before starting the next. All together around 40,000 single calculations have been done. To preserve disk space every output was compressed directly into gzip either by piping with the bash command gzip or by the implementation of gzip into the GZIPStream in Java BufferedReader/Writer.The complete GDB is more than 400 GB as gzip.

Enumeration

Graphs

The program Nauty from McKay was used to generate the connectivity tables for graphs, as found under http://cs.anu.edu.au/people/bdm/. The Nauty subprogram GENG was run up to 17 nodes for the generation of all possible graphs/geng -cd1D4 (Number of Nodes) The check was done for planarity of the graphs by PLANARG to avoid molecules with crossed bonds, e.g. Claus’ benzene./planarg

Hydrocarbons

The resulting output of GENG is a G6 string for a connection table, which was imported and converted to the corresponding hydrocarbons by exchanging the nodes with carbon atoms and the edges with single bonds. Hydrocarbons were filtered for desirable features (Table 1) because the majority of graphs includes 3- and 4-membered rings or multiple connected globular ring systems.

Skeletons

Each single bond was checked for the combinatorially introduction of double bonds (only four bonds per carbon are possible). Additionally every resulting double bond was checked for the combinatorially introduction of triple bonds. The resulting hydrocarbons were aromatized and dearomatized to avoid multiple copies of the same aromatic ring system. Unsaturated hydrocarbons were filtered for desirable features, e.g. the majority of unsaturated hydrocarbons includes allenes (Table 2).

CNO Molecules

Every monovalent, divalent, and trivalent carbon position was checked for the substitution with nitrogen following valency rules. Each monovalent and divalent carbon position was then checked for the substitution with oxygen. Before each exchange it was checked if the position is adjacent to a nitrogen or oxygen atom to avoid the generation of heteroatom heteroatom bonds and speeding up computation. Additionally it was checked before if the position is next to a sp atom, in which case no N or O atom was introduced. Symmetric positions were calculated only once. The resulting CNO-molecules were checked for desirable features to avoid unstable functional groups and to reduce the combinatorial explosion (Table 3). Each molecule was converted to unique SMILES strings to check for duplicates before storing the molecule.

Postprocessing for Oximes, Nitro, CF₃, Halogens, and Sulfur

Postprocessing for introducing additional diversity was performed as described in Table 4.

Shape Analysis

The shape analysis was adapted from Sauer and Schwarz (41) and was written in Java 1.6. SMILES were converted into 3D structure using CORINA. (46) The position of the molecule was expressed in a (x,y,z)-coordinate system defined by its principal axes. For each principal axis the moment of inertia was calculated using the general equation

Specified for each axis it yields

in which the squares of the radii around the axes are defined as r_x²= y²+ z², r_y² = x² + z², and r_z² = x² + y². The moments of inertia I_x, I_y, and I_z were then sorted in ascending order to yield I₁, I₂, and I₃. I₁ and I₂ were finally divided by the highest moment of inertia I₃ to yield the values P1 = I₁/I₃ and P2 = I₂/I₃.

The (P1,P2)-plane defines a two-dimensional triangular space with distinct boundaries, i.e. structures cannot be found outside the triangle. The triangle also has three distinct edges defining the different dimensionality of molecular shapes: The upper left edge of the triangle (0,1), the lower center edge (0.5,0.5), and the upper right edge (1,1) define 1D rodlike, 2D disklike, and 3D spherical structures, respectively.

Stereoisomer Counting

The CORINA command ./corina.lnx -i t=smiles -o t=sdf -d ori,stergen,rs | grep ˈ$$$$ˈ | wc −l was used to count stereoisomers.

Distribution

A 50 million random subset of GDB-17 and the leadlike and leadlike/no small ring fraction of this subset are freely available for download as a SMILES list from www.gdb.unibe.ch.

Author Information

ARTICLE SECTIONS

Jump To

Corresponding Author
- Jean-Louis Reymond - Department of Chemistry and Biochemistry, NCCR TransCure, University of Berne, Freiestrasse 3, 3012 Berne, Switzerland; Email: [email protected]
Authors
- Lars Ruddigkeit - Department of Chemistry and Biochemistry, NCCR TransCure, University of Berne, Freiestrasse 3, 3012 Berne, Switzerland
- Ruud van Deursen - Biomolecular Screening Facility, NCCR Chemical Biology, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland
- Lorenz C. Blum - Department of Chemistry and Biochemistry, NCCR TransCure, University of Berne, Freiestrasse 3, 3012 Berne, Switzerland
Notes
The authors declare no competing financial interest.

Acknowledgment

ARTICLE SECTIONS

Jump To

This work was supported financially by the University of Berne, the Swiss National Science Foundation, the NCCR TransCure, and the NCCR Chemical Biology.

References

ARTICLE SECTIONS

Jump To

This article references 47 other publications.

1
Lipkus, A. H.; Yuan, Q.; Lucas, K. A.; Funk, S. A.; Bartelt, W. F.; Schenck, R. J.; Trippe, A. J. Structural diversity of organic chemistry. A scaffold analysis of the CAS Registry J. Org. Chem. 2008, 73, 4443– 4451
[ACS Full Text ], [CAS], Google Scholar
1
Structural Diversity of Organic Chemistry. A Scaffold Analysis of the CAS Registry
Lipkus, Alan H.; Yuan, Qiong; Lucas, Karen A.; Funk, Susan A.; Bartelt, William F., III; Schenck, Roger J.; Trippe, Anthony J.
Journal of Organic Chemistry (2008), 73 (12), 4443-4451CODEN: JOCEAH; ISSN:0022-3263. (American Chemical Society)
The anal. of chem. diversity has become a topic of considerable interest in recent years. This interest has been stimulated by the challenge of discovering novel small-mol. pharmaceuticals. The development of technologies such as combinatorial synthesis and high-throughput screening has made it possible to explore drug-like regions of chem. space in relatively short times. Chem. space is vast and the problem of selecting which region of that space to explore remains a key issue in drug discovery. By analyzing the scaffold content of the CAS Registry, the authors attempt to characterize in a comprehensive way the structural diversity of org. chem. The scaffold of a mol. is taken to be its framework, defined as all its ring systems and all the linkers that connect them. Framework data from more than 24 million org. compds. are analyzed. The distribution of frameworks among compds. is found to be top-heavy, i.e., a small percentage of frameworks occurs in a large percentage of compds. When frameworks are analyzed at the graph level, an even more top-heavy distribution is found: half of the compds. can be described by only 143 framework shapes. The most significant finding is that the framework distribution conforms almost exactly to a power law. This suggests that the more often a framework has been used as the basis for a compd., the more likely it is to be used in another compd. This may be explained by the cost of synthesis: making a new deriv. of a framework is probably less costly if many other derivs. are known. The authors believe that this power law is evidence that the minimization of synthetic cost has been a key factor in shaping the known universe of org. chem.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXmtlylu7c%253D&md5=c03682d3a3fab0b0eec7a36ef1ecc778
2
ACS NEWS Chem. Eng. News 2011, 89, 38
Google Scholar
There is no corresponding record for this reference.
3
Bleicher, K. H.; Bohm, H. J.; Muller, K.; Alanine, A. I. Hit and lead generation: Beyond high-throughput screening Nat. Rev. Drug Discovery 2003, 2, 369– 378
[Crossref], [PubMed], [CAS], Google Scholar
3
A guide to drug discovery: Hit and lead generation: beyond high-throughput screening
Bleicher, Konrad H.; Boehm, Hans-Joachim; Mueller, Klaus; Alanine, Alexander I.
Nature Reviews Drug Discovery (2003), 2 (5), 369-378CODEN: NRDDAG; ISSN:1474-1776. (Nature Publishing Group)
A review. The identification of small-mol. modulators of protein function, and the process of transforming these into high-content lead series, are key activities in modern drug discovery. The decisions taken during this process have far-reaching consequences for success later in lead optimization and even more crucially in clin. development. Recently, there has been an increased focus on these activities due to escalating downstream costs resulting from high clin. failure rates. In addn., the vast emerging opportunities from efforts in functional genomics and proteomics demands a departure from the linear process of identification, evaluation and refinement activities towards a more integrated parallel process. This calls for flexible, fast and cost-effective strategies to meet the demands of producing high-content lead series with improved prospects for clin. success.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXjslamtb8%253D&md5=b54e55c1f0d34fbd172c8aa9b996591d
4
Schreiber, S. L. Small molecules: the missing link in the central dogma Nat. Chem. Biol. 2005, 1, 64– 66
[Crossref], [PubMed], [CAS], Google Scholar
4
Small molecules: the missing link in the central dogma
Schreiber, Stuart L.
Nature Chemical Biology (2005), 1 (2), 64-66CODEN: NCBABT; ISSN:1552-4450. (Nature Publishing Group)
There is no expanded citation for this reference.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXls1Oms7Y%253D&md5=d4a54994c0d933602d9d02747e5e1ddc
5
Mayr, L. M.; Bojanic, D. Novel trends in high-throughput screening Curr. Opin. Pharmacol. 2009, 9, 580– 588
[Crossref], [PubMed], [CAS], Google Scholar
5
Novel trends in high-throughput screening
Mayr, Lorenz M.; Bojanic, Dejan
Current Opinion in Pharmacology (2009), 9 (5), 580-588CODEN: COPUBK; ISSN:1471-4892. (Elsevier B.V.)
A review. High-throughput screening (HTS) is a well-established process for lead discovery in Pharma and Biotech companies and is now also being used for basic and applied research in academia. It comprises the screening of large chem. libraries for activity against biol. targets via the use of automation, miniaturized assays and large-scale data anal. Since its first advent in the early to mid 1990s, the field of HTS has seen not only a continuous change in technol. and processes, but also an adaptation to various needs in lead discovery. HTS has now evolved into a mature discipline that is a crucial source of chem. starting points for drug discovery. Whereas in previous years much emphasis has been put on a steady increase in screening capacity (quant. increase') via automation and miniaturization, the past years have seen a much greater emphasis on content and quality (qual. increase'). Today, many experts in the field see HTS at a crossroad with the need to decide on either higher throughput/more experimentation or a greater focus on assays of greater physiol. relevance, both of which may lead to higher productivity in pharmaceutical R&D. In this paper, we describe the development of HTS over the past decade and point out our own ideas for future directions of HTS in biomedical research. We predict that the trend toward further miniaturization will slow down with the balanced implementation of 384 well, 1536 well, and 384 low vol. well plates. Furthermore, we envisage that there will be much more emphasis on rigorous assay and chem. characterization, particularly considering that novel and more difficult target classes will be pursued. In recent years we have witnessed a clear trend in the drug discovery community toward rigorous hit validation by the use of orthogonal readout technologies, label free and biophys. methodologies. We also see a trend toward a more flexible use of the various screening approaches in lead discovery, i.e., the use of both full deck compd. screening as well as the use of focused screening and iterative screening approaches. Moreover, we expect greater usage of target identification strategies downstream of phenotypic screening and the more effective implementation of affinity selection technologies as a result of advances in chem. diversity methodologies. We predict that, ultimately, each hit finding strategy will be much more project-related, tailor-made, and better integrated into the broader drug discovery efforts.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXht1WntLnL&md5=50284fdd9f4d610e81fb442b78b7b69e
6
Renner, S.; Popov, M.; Schuffenhauer, A.; Roth, H. J.; Breitenstein, W.; Marzinzik, A.; Lewis, I.; Krastel, P.; Nigsch, F.; Jenkins, J.; Jacoby, E. Recent trends and observations in the design of high-quality screening collections Future Med. Chem 2011, 3, 751– 766
[Crossref], [PubMed], [CAS], Google Scholar
6
Recent trends and observations in the design of high-quality screening collections
Renner, Steffen; Popov, Maxim; Schuffenhauer, Ansgar; Roth, Hans-Joerg; Breitenstein, Werner; Marzinzik, Andreas; Lewis, Ian; Krastel, Philipp; Nigsch, Florian; Jenkins, Jeremy; Jacoby, Edgar
Future Medicinal Chemistry (2011), 3 (6), 751-766CODEN: FMCUA7; ISSN:1756-8919. (Future Science Ltd.)
A review. The design of a high-quality screening collection is of utmost importance for the early drug-discovery process and provides, in combination with high-quality assay systems, the foundation of future discoveries. Herein, we review recent trends and observations to successfully expand the access to bioactive chem. space, including the feedback from hit assessment interviews of high-throughput screening campaigns; recent successes with chemogenomics target family approaches, the identification of new relevant target/domain families, diversity-oriented synthesis and new emerging compd. classes, and non-classical approaches, such as fragment-based screening and DNA-encoded chem. libraries. The role of in silico library design approaches are emphasized.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXlvFGisLY%253D&md5=f876f165f937c9792c989053492e59f5
7
Kola, I.; Landis, J. Can the pharmaceutical industry reduce attrition rates? Nat. Rev. Drug Discovery 2004, 3, 711– 715
[Crossref], [PubMed], [CAS], Google Scholar
7
Opinion: Can the pharmaceutical industry reduce attrition rates?
Kola, Ismail; Landis, John
Nature Reviews Drug Discovery (2004), 3 (8), 711-716CODEN: NRDDAG; ISSN:1474-1776. (Nature Publishing Group)
The pharmaceutical industry faces considerable challenges, both politically and fiscally. Politically, governments around the world are trying to contain costs and, as health care budgets constitute a very significant part of governmental spending, these costs are the subject of intense scrutiny. In the United States, drug costs are also the subject of intense political discourse. This article deals with the fiscal pressures that face the industry from the perspective of R&D. What impinges on productivity How can we improve current reduced R&D productivity.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXmtVOhtLs%253D&md5=f9025c13a1506f607aaf68415570ed01
8
Hann, M. M. Molecular obesity, potency and other addictions in drug discovery MedChemComm 2011, 2, 349– 355
[Crossref], [CAS], Google Scholar
8
Molecular obesity, potency and other addictions in drug discovery
Hann, Michael M.
MedChemComm (2011), 2 (5), 349-355CODEN: MCCEAY; ISSN:2040-2503. (Royal Society of Chemistry)
A review. Despite the increase in global biol. and chem. knowledge the discovery of effective and safe new drugs seems to become harder rather than easier. Some of this challenge is due to increasing demands for safety and novelty, but some of the risk involved in this should be controllable if we had more effectively learned from our failures. This perspective reflects on some of the learnings of recent years in relation to the causes of attrition. The term Mol. Obesity is introduced to describe our tendency to build potency into mols. by the inappropriate use of lipophilicity which leads to the premature demise of drug candidates.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXls1Kht7g%253D&md5=954870db4b1d9e846612c3bb256582f7
9
Schneider, G.; Fechner, U. Computer-based de novo design of drug-like molecules Nat. Rev. Drug Discovery 2005, 4, 649– 663
[Crossref], [PubMed], [CAS], Google Scholar
9
Computer-based de novo design of drug-like molecules
Schneider, Gisbert; Fechner, Uli
Nature Reviews Drug Discovery (2005), 4 (8), 649-663CODEN: NRDDAG; ISSN:1474-1776. (Nature Publishing Group)
A review with refs. Ever since the first automated de novo design techniques were conceived only 15 years ago, the computer-based design of hit and lead structure candidates has emerged as a complementary approach to high-throughput screening. Although many challenges remain, de novo design supports drug discovery projects by generating novel pharmaceutically active agents with desired properties in a cost- and time-efficient manner. In this review, we outline the various design concepts and highlight current developments in computer-based de novo design.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXmvVOqtro%253D&md5=a30dbc58ed81e0b7fe3f7d41a668e9ac
10
Jorgensen, W. L. Efficient drug lead discovery and optimization Acc. Chem. Res. 2009, 42, 724– 733
[ACS Full Text ], [CAS], Google Scholar
10
Efficient Drug Lead Discovery and Optimization
Jorgensen, William L.
Accounts of Chemical Research (2009), 42 (6), 724-733CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
A review. During the 1980s, advances in the abilities to perform computer simulations of chem. and biomol. systems and to calc. free energy changes led to the expectation that such methodol. would soon show great utility for guiding mol. design. Important potential applications included design of selective receptors, catalysts, and regulators of biol. function including enzyme inhibitors. This time also saw the rise of high-throughput screening and combinatorial chem. along with complementary computational methods for de novo design and virtual screening including docking. These technologies appeared poised to deliver diverse lead compds. for any biol. target. As with many technol. advances, realization of the expectations required significant addnl. effort and time. However, as summarized here, striking success has now been achieved for computer-aided drug lead generation and optimization. De novo design using both mol. growing and docking are illustrated for lead generation, and lead optimization features free energy perturbation calcns. in conjunction with Monte Carlo statistical mechanics simulations for protein-inhibitor complexes in aq. soln. The specific applications are to the discovery of non-nucleoside inhibitors of HIV reverse transcriptase (HIV-RT) and inhibitors of the binding of the proinflammatory cytokine MIF to its receptor CD74. A std. protocol is presented that includes scans for possible addns. of small substituents to a mol. core, interchange of heterocycles, and focused optimization of substituents at one site. Initial leads with activities at low-micromolar concns. have been advanced rapidly to low-nanomolar inhibitors.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXjsFentr4%253D&md5=9641587769701c6541b38e18fa05538a
11
Reymond, J. L.; Van Deursen, R.; Blum, L. C.; Ruddigkeit, L. Chemical space as a source for new drugs MedChemComm 2010, 1, 30– 38
[Crossref], [CAS], Google Scholar
11
Chemical space as a source for new drugs
Reymond, Jean-Louis; van Deursen, Ruud; Blum, Lorenz C.; Ruddigkeit, Lars
MedChemComm (2010), 1 (1), 30-38CODEN: MCCEAY; ISSN:2040-2503. (Royal Society of Chemistry)
A review. The chem. space is the ensemble of all possible mols., which is believed to contain at least 1060 org. mols. below 500 Da of possible interest for drug discovery. This review summarizes the development of the chem. space concept from enumerating acyclic hydrocarbons in the 1800's to the recent assembly of the chem. universe database GDB. Chem. space travel algorithms can be used to explore defined regions of chem. space by generating focused virtual libraries. Maps of the chem. space are produced from property spaces visualized by principal component anal. or by self-organizing maps, and from structural analyses such as the scaffold-tree or the MQN-system. Virtual screening of virtual chem. space followed by synthesis and testing of the best hits leads to the discovery of new drug mols.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhtVeju7rF&md5=579428edc79f706101ab58c505052c72
12
Hartenfeller, M.; Schneider, G. De novo drug design Methods Mol. Biol. 2011, 672, 299– 323
[Crossref], [PubMed], [CAS], Google Scholar
12
De novo drug design
Hartenfeller, Markus; Schneider, Gisbert
Methods in Molecular Biology (New York, NY, United States) (2011), 672 (Chemoinformatics and Computational Chemical Biology), 299-323CODEN: MMBIED; ISSN:1064-3745. (Springer)
A review. Computer-assisted mol. design supports drug discovery by suggesting novel chemotypes and compd. modifications for lead structure optimization. While the aspect of synthetic feasibility of the automatically designed compds. has been neglected for a long time, we are currently witnessing an increased interest in this topic. Here, we review state-of-the-art software for de novo drug design with a special emphasis on fragment-based techniques that generate druglike, synthetically accessible compds. The importance of scoring functions that can be used to predict compd. reactivity and potency is highlighted, and several promising solns. are discussed. Recent practical validation studies are presented that have already demonstrated that rule-based fragment assembly can result in novel synthesizable compds. with druglike properties and a desired biol. activity.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhtlejsr7L&md5=3a30147da9187fa4895ba807cefed82a
13
Klebe, G. Virtual ligand screening: strategies, perspectives and limitations Drug Discovery Today 2006, 11, 580– 594
[Crossref], [PubMed], [CAS], Google Scholar
13
Virtual ligand screening: strategies, perspectives and limitations
Klebe, Gerhard
Drug Discovery Today (2006), 11 (13 & 14), 580-594CODEN: DDTOFS; ISSN:1359-6446. (Elsevier B.V.)
A review. In contrast to high-throughput screening, in virtual ligand screening (VS), compds. are selected using computer programs to predict their binding to a target receptor. A key prerequisite is knowledge about the spatial and energetic criteria responsible for protein-ligand binding. The concepts and prerequisites to perform VS are summarized here, and explanations are sought for the enduring limitations of the technol. Target selection, anal. and prepn. are discussed, as well as considerations about the compilation of candidate ligand libraries. The tools and strategies of a VS campaign, and the accuracy of scoring and ranking of the results, are also considered.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD28XlvFGqtLo%253D&md5=5adfd125d48082238a6c5ad0e8343c59
14
Kolb, P.; Ferreira, R. S.; Irwin, J. J.; Shoichet, B. K. Docking and chemoinformatic screens for new ligands and targets Curr. Opin. Biotechnol. 2009, 20, 429– 36
[Crossref], [PubMed], [CAS], Google Scholar
14
Docking and chemoinformatic screens for new ligands and targets
Kolb, Peter; Ferreira, Rafaela S.; Irwin, John J.; Shoichet, Brian K.
Current Opinion in Biotechnology (2009), 20 (4), 429-436CODEN: CUOBE3; ISSN:0958-1669. (Elsevier B.V.)
A review. Computer-based docking screens are now widely used to discover new ligands for targets of known structure; in the last two years alone, the discovery of ligands for more than 20 proteins has been reported. Recently, investigators have also turned to predicting new substrates for enzymes of unknown function, taking docking in a wholly new direction. Increasingly, the hit rates, the true-positives, and the false-positives from the docking screens are being compared to those from empirical, high-throughput screens, revealing the strengths, weaknesses, and complementarities of both techniques. The recent efflorescence of GPCR structures has made these quintessential drug targets available to structure-based approaches. Consistent with their druggability', the docking screens have returned high hit rates and potent mols. Finally, in the last several years, an approach almost exactly opposite to docking has also appeared; this pharmacol. network approach begins not with the structure of the target but rather those of drug mols. and asks, given a pattern of chem. in the ligands, what targets may a particular drug bind to. This method, which returns to an older, pharmacol. logic, has been surprisingly successful in predicting new off-targets' for established drugs.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXht1OrtLbL&md5=4b6b2457a735e18be5780d9299ebffe6
15
Geppert, H.; Vogt, M.; Bajorath, J. Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation J. Chem. Inf. Model. 2010, 50, 205– 216
[ACS Full Text ], [CAS], Google Scholar
15
Current Trends in Ligand-Based Virtual Screening: Molecular Representations, Data Mining Methods, New Application Areas, and Performance Evaluation
Geppert, Hanna; Vogt, Martin; Bajorath, Jurgen
Journal of Chemical Information and Modeling (2010), 50 (2), 205-216CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
There is no expanded citation for this reference.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXnvFWmsA%253D%253D&md5=aeb674ef7c6c93711eb21452732e21d8
16
Cayley, E. Ueber die analytischen Figuren, welche in der Mathematik Bäume genannt werden und ihre Anwendung auf die Theorie chemischer Verbindungen Chem. Ber. 1875, 8, 1056– 1059
[Crossref], Google Scholar
There is no corresponding record for this reference.
17
Lederberg, J.; Sutherland, G. L.; Buchanan, B. G.; Feigenbaum, E. A.; Robertson, A. V.; Duffield, A. M.; Djerassi, C. Applications of artificial intelligence for chemical inference. I. Number of possible organic compounds. Acyclic structures containing carbon, hydrogen, oxygen, and nitrogen J. Am. Chem. Soc. 1969, 91, 2973– 2976
[ACS Full Text ], Google Scholar
There is no corresponding record for this reference.
18
Steinbeck, C. Recent developments in automated structure elucidation of natural products Nat. Prod. Rep. 2004, 21, 512– 518
[Crossref], [PubMed], [CAS], Google Scholar
18
Recent developments in automated structure elucidation of natural products
Steinbeck, Christoph
Natural Product Reports (2004), 21 (4), 512-518CODEN: NPRRDF; ISSN:0265-0568. (Royal Society of Chemistry)
A review. Advancements in the field of computer-assisted structure elucidation (CASE) of natural products achieved in the past five years are discussed. This process starts with a dereplication procedure, supported by structure-spectrum databases. Both com. and free products are available to support the procedure. A no. of new programs, as well as advancements in existing ones, are presented. Finally, the option to validate the result by an independent procedure, a high quality ab initio quantum mech. calcn., is discussed.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXnsVKltr0%253D&md5=7d7486d14d46798ede0aa853faf013f9
19
Reymond, J. L.; Ruddigkeit, L.; Blum, L. C.; Van Deursen, R. The enumeration of chemical space Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 717– 733
[Crossref], [CAS], Google Scholar
19
The enumeration of chemical space
Reymond, Jean-Louis; Ruddigkeit, Lars; Blum, Lorenz; van Deursen, Ruud
Wiley Interdisciplinary Reviews: Computational Molecular Science (2012), 2 (5), 717-733CODEN: WIRCAH; ISSN:1759-0884. (Wiley-Blackwell)
A review. In the field of medicinal chem., the chem. space describes the ensemble of all org. mols. to be considered when searching for new drugs (estd. >1060 mols.), as well as the property spaces in which these mols. are placed for the sake of describing them. Mols. can be enumerated computationally by the millions, which was first undertaken in the field of computer-aided structure elucidation. Scoring the enumerated virtual libraries by virtual screening has recently become an attractive strategy to prioritize compds. for synthesis and testing. Enumeration methods include combinatorial linking of fragments, genetic algorithms based on cycles of enumeration and selection by ligand-based or target-based scoring functions, and exhaustive enumeration from first principles. The chem. space of mols. following simple rules of chem. stability and synthetic feasibility has been enumerated up to 13 atoms of C, N, O, Cl, S, forming the GDB-13 database with 977 million structures. The database has been organized in a 42-dimensional chem. space using mol. quantum nos. (MQN) as descriptors, which can be visualized by projection in two dimensions by principal component anal.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhsFCmsr%252FM&md5=cda6dfd2d048e668455f52ee884b718a
20
Fink, T.; Bruggesser, H.; Reymond, J. L. Virtual exploration of the small-molecule chemical universe below 160 Da Angew. Chem., Int. Ed. Engl. 2005, 44, 1504– 1508
[Crossref], [PubMed], Google Scholar
There is no corresponding record for this reference.
21
Fink, T.; Reymond, J. L. Virtual exploration of the chemical universe up to 11 atoms of C, N, O, F: assembly of 26.4 million structures (110.9 million stereoisomers) and analysis for new ring systems, stereochemistry, physicochemical properties, compound classes, and drug discovery J. Chem. Inf. Model. 2007, 47, 342– 353
[ACS Full Text ], [CAS], Google Scholar
21
Virtual Exploration of the Chemical Universe up to 11 Atoms of C, N, O, F: Assembly of 26.4 Million Structures (110.9 Million Stereoisomers) and Analysis for New Ring Systems, Stereochemistry, Physicochemical Properties, Compound Classes, and Drug Discovery
Fink, Tobias; Reymond, Jean-Louis
Journal of Chemical Information and Modeling (2007), 47 (2), 342-353CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
All mols. of up to 11 atoms of C, N, O, and F possible under consideration of simple valency, chem. stability, and synthetic feasibility rules were generated and collected in a database (GDB). GDB contains 26.4 million mols. (110.9 million stereoisomers), including three- and four-membered rings and triple bonds. By comparison, only 63 857 compds. of up to 11 atoms were found in public databases (a combination of PubChem, ChemACX, ChemSCX, NCI open database, and the Merck Index). A total of 538 of the 1208 ring systems in GDB are currently unknown in the CAS Registry and Beilstein databases in any carbon/heteroatom/multiple-bond combination or as a substructure. Over 70% of GDB mols. are chiral. Because of their small size, all compds. obey Lipinski's bioavailability rule. A total of 13.2 million compds. also follow Congreve's "Rule of 3" for lead-likeness. A Kohonen map trained with autocorrelation descriptors organizes GDB according to compd. classes and shows that leadlike compds. are most abundant in chiral regions of fused carbocycles and fused heterocycles. The projection of known compds. into this map indicates large uncharted areas of chem. space. The potential of GDB for drug discovery is illustrated by virtual screening for kinase inhibitors, G-protein coupled receptor ligands, and ion-channel modulators. The database is available from the author's Web page.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2sXhtVens7k%253D&md5=fe97c30ee8269de1e889648f4818f42f
22
Blum, L. C.; Reymond, J. L. 970 million druglike small molecules for virtual screening in the chemical universe database GDB-13 J. Am. Chem. Soc. 2009, 131, 8732– 8733
[ACS Full Text ], [CAS], Google Scholar
22
970 Million Druglike Small Molecules for Virtual Screening in the Chemical Universe Database GDB-13
Blum, Lorenz C.; Reymond, Jean-Louis
Journal of the American Chemical Society (2009), 131 (25), 8732-8733CODEN: JACSAT; ISSN:0002-7863. (American Chemical Society)
GDB-13 enumerates small org. mols. contg. up to 13 atoms of C, N, O, S, and Cl following simple chem. stability and synthetic feasibility rules. With 977 468 314 structures, GDB-13 is the largest publicly available small org. mol. database to date.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXmvFWru7k%253D&md5=22c200e887a6480b19a73852ca0a3435
23
Blum, L. C.; van Deursen, R.; Reymond, J. L. Visualisation and subsets of the chemical universe database GDB-13 for virtual screening J. Comput.-Aided Mol. Des. 2011, 25, 637– 647
[Crossref], [PubMed], [CAS], Google Scholar
23
Visualisation and subsets of the chemical universe database GDB-13 for virtual screening
Blum, Lorenz C.; Deursen, Ruud; Reymond, Jean-Louis
Journal of Computer-Aided Molecular Design (2011), 25 (7), 637-647CODEN: JCADEQ; ISSN:0920-654X. (Springer)
The chem. universe database GDB-13, which enumerates 977 million org. mols. up to 13 atoms of C, N, O, S and Cl following simple chem. stability and synthetic feasibility rules, represents a vast reservoir for new fragments. GDB-13 was classified using the MQN-system discussed in the preceding paper for the anal. of PubChem fragments. Two hundred and fifty-five subsets of GDB-13 were generated by the combinatorial use of eight restrictive criteria, including fragment-like ("rule of three") and scaffold-like (no acyclic carbon atoms) filters. Virtual screening for analogs of 15 com. drugs of 13 non-hydrogen atoms or less shows that retrieving MQN-neighbors of a query mol. from GDB-13 or its subsets provides on av. a 38-fold enrichment in structural analogs (Daylight-type substructure fingerprint Tanimoto TSF > 0.7), and a 75-fold enrichment in shape-similar analogs (ROCS TanimotoCombo score > 1.4). An MQN-searchable version of GDB-13 is provided at www.ghb.unibe.ch.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhtVSqtbbN&md5=45f426cafa64b3bb06fc11e68a1b4022
24
Nguyen, K. T.; Syed, S.; Urwyler, S.; Bertrand, S.; Bertrand, D.; Reymond, J. L. Discovery of NMDA glycine site inhibitors from the chemical universe database GDB ChemMedChem 2008, 3, 1520– 1524
[Crossref], [PubMed], [CAS], Google Scholar
24
Discovery of NMDA glycine site inhibitors from the chemical universe database GDB
Nguyen, Kong Thong; Syed, Salahuddin; Urwyler, Stephan; Bertrand, Sonia; Bertrand, Daniel; Reymond, Jean-Louis
ChemMedChem (2008), 3 (10), 1520-1524CODEN: CHEMGX; ISSN:1860-7179. (Wiley-VCH Verlag GmbH & Co. KGaA)
Using virtual screening tools, promising small mol. ligands of the NMDA receptor glycine site are identified for subsequent synthesis.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXhtlOgt73P&md5=d240511779046f0794b75e35f52c42cc
25
Nguyen, K. T.; Luethi, E.; Syed, S.; Urwyler, S.; Bertrand, S.; Bertrand, D.; Reymond, J. L. 3-(aminomethyl)piperazine-2,5-dione as a novel NMDA glycine site inhibitor from the chemical universe database GDB Bioorg. Med. Chem. Lett. 2009, 19, 3832– 3835
[Crossref], [PubMed], Google Scholar
There is no corresponding record for this reference.
26
Garcia-Delgado, N.; Bertrand, S.; Nguyen, K. T.; van Deursen, R.; Bertrand, D.; Reymond, J.-L. Exploring a7-nicotinic receptor ligand diversity by scaffold enumeration from the Chemical Universe Database GDB ACS Med. Chem. Lett. 2010, 1, 422– 426
[ACS Full Text ], Google Scholar
There is no corresponding record for this reference.
27
Luethi, E.; Nguyen, K. T.; Burzle, M.; Blum, L. C.; Suzuki, Y.; Hediger, M.; Reymond, J. L. Identification of selective norbornane-type aspartate analogue inhibitors of the glutamate transporter 1 (GLT-1) from the chemical universe generated database (GDB) J. Med. Chem. 2010, 53, 7236– 7250
[ACS Full Text ], [CAS], Google Scholar
27
Identification of Selective Norbornane-Type Aspartate Analogue Inhibitors of the Glutamate Transporter 1 (GLT-1) from the Chemical Universe Generated Database (GDB)
Luethi, Erika; Nguyen, Kong T.; Burzle, Marc; Blum, Lorenz C.; Suzuki, Yoshiro; Hediger, Matthias; Reymond, Jean-Louis
Journal of Medicinal Chemistry (2010), 53 (19), 7236-7250CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
A variety of conformationally constrained aspartate and glutamate analogs inhibit the glutamate transporter 1 (GLT-1, also known as EAAT2). To expand the search for such analogs, a virtual library of aliph. aspartate and glutamate analogs was generated starting from the chem. universe database GDB-11, which contains 26.4 million possible mols. up to 11 atoms of C, N, O, F, resulting in 101026 aspartate analogs and 151285 glutamate analogs. Virtual screening was realized by high-throughput docking to the glutamate binding site of the glutamate transporter homolog from Pyrococcus horikoshii (PDB code: 1XFH) using Autodock. Norbornane-type aspartate analogs were selected from the top-scoring virtual hits and synthesized. Testing and optimization led to the identification of (1R*,2R*,3S*,4R*,6R*)-2-amino-6-phenethyl-bicyclo[2.2.1]heptane-2,3-dicarboxylic acid (I) as a new inhibitor of GLT-1 with IC50 = 1.4 μM against GLT-1 and no inhibition of the related transporter EAAC1. The systematic diversification of known ligands by enumeration with help of GDB followed by virtual screening, synthesis, and testing as exemplified here provides a general strategy for drug discovery.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhtFWlsLvF&md5=3834d28f487e11ead1a8a17698587935
28
Blum, L. C.; van Deursen, R.; Bertrand, S.; Mayer, M.; Burgi, J. J.; Bertrand, D.; Reymond, J. L. Discovery of alpha7-nicotinic receptor ligands by virtual screening of the Chemical Universe Database GDB-13 J. Chem. Inf. Model. 2011, 51, 3105– 3112
[ACS Full Text ], [CAS], Google Scholar
28
Discovery of α7-Nicotinic Receptor Ligands by Virtual Screening of the Chemical Universe Database GDB-13
Blum, Lorenz C.; van Deursen, Ruud; Bertrand, Sonia; Mayer, Milena; Burgi, Justus J.; Bertrand, Daniel; Reymond, Jean-Louis
Journal of Chemical Information and Modeling (2011), 51 (12), 3105-3112CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
The chem. universe database GDB-13 enumerates 977 million org. mols. up to 13 atoms of C, N, O, Cl, and S that are virtually possible following simple rules for chem. stability and synthetic feasibility. Analogs of nicotine were identified in GDB-13 using the city-block distance in MQN-space (CBDMQN) as a similarity measure, combined with a restriction eliminating problematic structural elements. The search was carried out with a Web browser available at www.gdb.unibe.ch. This virtual screening procedure selected 31 504 analogs of nicotine from GDB-13, from which 48 were known nicotinic ligands reported in Chembl. An addnl. 60 virtual screening hits were purchased and tested for modulation of the acetylcholine signal at the human α7 nAChR expressed in Xenopus oocytes, which led to the identification of three previously unknown inhibitors. These expts. demonstrate for the first time the use of GDB-13 for ligand discovery.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhsVGru7vK&md5=c176859de6f18cc54b2f25eee42ab79a
29
Brethous, L.; Garcia-Delgado, N.; Schwartz, J.; Bertrand, S.; Bertrand, D.; Reymond, J. L. Synthesis and nicotinic receptor activity of chemical space analogues of N-(3R)-1-azabicyclo[2.2.2]oct-3-yl-4-chlorobenzamide (PNU-282,987) and 1,4-diazabicyclo[3.2.2]nonane-4-carboxylic acid 4-bromophenyl ester (SSR180711) J. Med. Chem. 2012, 55, 4605– 4618
[ACS Full Text ], Google Scholar
There is no corresponding record for this reference.
30
Reymond, J. L.; Awale, M. Exploring chemical space for drug discovery using the Chemical Universe Database ACS Chem. Neurosci. 2012, 3, 649– 657
[ACS Full Text ], [CAS], Google Scholar
30
Exploring Chemical Space for Drug Discovery Using the Chemical Universe Database
Reymond, Jean-Louis; Awale, Mahendra
ACS Chemical Neuroscience (2012), 3 (9), 649-657CODEN: ACNCDM; ISSN:1948-7193. (American Chemical Society)
Herein we review our recent efforts in searching for bioactive ligands by enumeration and virtual screening of the unknown chem. space of small mols. Enumeration from first principles shows that almost all small mols. (>99.9%) have never been synthesized and are still available to be prepd. and tested. We discuss open access sources of mols., the classification and representation of chem. space using mol. quantum nos. (MQN), its exhaustive enumeration in form of the chem. universe generated databases (GDB), and examples of using these databases for prospective drug discovery. MQN-searchable GDB, PubChem, and DrugBank are freely accessible at www.gdb.unibe.ch.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XmtValu70%253D&md5=bbbe3c0931328f3796ce999189374864
31
Foloppe, N. The benefits of constructing leads from fragment hits Future Med. Chem. 2011, 3, 1111– 1115
[Crossref], [PubMed], [CAS], Google Scholar
31
The benefits of constructing leads from fragment hits
Foloppe, N.
Future Medicinal Chemistry (2011), 3 (9), 1111-1115CODEN: FMCUA7; ISSN:1756-8919. (Future Science Ltd.)
A review. Fragments'' refer to particularly small mol. starting points in medicinal chem. The small size of fragments requires adapted techniques for their screening and subsequent elaboration. The detection of the weak binding affinity of fragments for their target, and assocd. screening issues, have been debated at length. Since it is now clear that fragments can be developed into clin. candidates, the discussion is shifting to the design of good-quality lead compds. from fragment hits. The increasing ability to control and tailor this construction process highlights the potential benefits of fragment-based drug discovery.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXpsFeisLY%253D&md5=915652874047d19b21987cccf52c0df7
32
Teague, S. J.; Davis, A. M.; Leeson, P. D.; Oprea, T. The design of leadlike combinatorial libraries Angew. Chem., Int. Ed. Engl. 1999, 38, 3743– 3748
[Crossref], Google Scholar
There is no corresponding record for this reference.
33
Wang, Y.; Xiao, J.; Suzek, T. O.; Zhang, J.; Wang, J.; Bryant, S. H. PubChem: a public information system for analyzing bioactivities of small molecules Nucleic Acids Res. 2009, 37, W623– W633
[Crossref], [PubMed], [CAS], Google Scholar
33
PubChem: a public information system for analyzing bioactivities of small molecules
Wang, Yanli; Xiao, Jewen; Suzek, Tugba O.; Zhang, Jian; Wang, Jiyao; Bryant, Stephen H.
Nucleic Acids Research (2009), 37 (Web Server), W623-W633CODEN: NARHAD; ISSN:0305-1048. (Oxford University Press)
PubChem (http://pubchem.ncbi.nlm.nih.gov) is a public repository for biol. properties of small mols. hosted by the US National Institutes of Health (NIH). PubChem BioAssay database currently contains biol. test results for more than 700 000 compds. The goal of PubChem is to make this information easily accessible to biomedical researchers. In this work, we present a set of web servers to facilitate and optimize the utility of biol. activity information within PubChem. These web-based services provide tools for rapid data retrieval, integration and comparison of biol. screening results, exploratory structure-activity anal., and target selectivity examn. This article reviews these bioactivity anal. tools and discusses their uses. Most of the tools described in this work can be directly accessed at http://pubchem.ncbi.nlm.nih.gov/assay/. URLs for accessing other tools described in this work are specified individually.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXosFSktL8%253D&md5=11d50a1ff4d9b353728e1e03ee2e33ca
34
Gaulton, A.; Bellis, L. J.; Bento, A. P.; Chambers, J.; Davies, M.; Hersey, A.; Light, Y.; McGlinchey, S.; Michalovich, D.; Al-Lazikani, B.; Overington, J. P. ChEMBL: a large-scale bioactivity database for drug discovery Nucleic Acids Res. 2012, 40, D1100– D1107
[Crossref], [PubMed], [CAS], Google Scholar
34
ChEMBL: a large-scale bioactivity database for drug discovery
Gaulton, Anna; Bellis, Louisa J.; Bento, A. Patricia; Chambers, Jon; Davies, Mark; Hersey, Anne; Light, Yvonne; McGlinchey, Shaun; Michalovich, David; Al-Lazikani, Bissan; Overington, John P.
Nucleic Acids Research (2012), 40 (D1), D1100-D1107CODEN: NARHAD; ISSN:0305-1048. (Oxford University Press)
ChEMBL is an Open Data database contg. binding, functional and ADMET information for a large no. of drug-like bioactive compds. These data are manually abstracted from the primary published literature on a regular basis, then further curated and standardized to maximize their quality and utility across a wide range of chem. biol. and drug-discovery research problems. Currently, the database contains 5.4 million bioactivity measurements for more than 1 million compds. and 5200 protein targets. Access is available through a web-based interface, data downloads and web services at: https://www.ebi.ac.uk/chembldb.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhs12htbjN&md5=aedf7793e1ca54b6a4fa272ea3ef7d0e
35
Knox, C.; Law, V.; Jewison, T.; Liu, P.; Ly, S.; Frolkis, A.; Pon, A.; Banco, K.; Mak, C.; Neveu, V.; Djoumbou, Y.; Eisner, R.; Guo, A. C.; Wishart, D. S. DrugBank 3.0: a comprehensive resource for ‘Omics’ research on drugs Nucleic Acids Res. 2011, 39, D1035– D1041
[Crossref], Google Scholar
There is no corresponding record for this reference.
36
McKay, B. D. Practical graph isomorphism Congressus Numerantium 1981, 30, 45– 87
Google Scholar
There is no corresponding record for this reference.
37
Rishton, G. M. Reactive compounds and in vitro false positives in HTS Drug Discovery Today 1997, 2, 382– 384
[Crossref], [CAS], Google Scholar
37
Reactive compounds and in vitro false positives in HTS
Rishton, Gilbert M.
Drug Discovery Today (1997), 2 (9), 382-384CODEN: DDTOFS; ISSN:1359-6446. (Elsevier)
A review without refs. An important component of the successful high-throughput screening (HTS) strategy in drug discovery is the ability to assess HTS structure-activity data, and to distinguish between promising drug leads and the many useless false positives that can plague screening efforts. The author discusses simple chem. guidelines for the evaluation of "positives" in biochem. screens, with the aim of selecting stable, non-covalent binders (ligands) and eliminating protein-reactive compds. (reagents) from consideration as drug leads at an early stage.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK2sXlvFeksL4%253D&md5=b43a00783098a324ab3dc4d5f3704013
38
Rishton, G. M. Nonleadlikeness and leadlikeness in biochemical screening Drug Discovery Today 2003, 8, 86– 96
[Crossref], [PubMed], [CAS], Google Scholar
38
Nonleadlikeness and leadlikeness in biochemical screening
Rishton Gilbert M
Drug discovery today (2003), 8 (2), 86-96 ISSN:1359-6446.
Biochemical assays have largely supplanted functional biological assays as drug screening tools in the early stages of drug discovery. The de-selection of compounds that are 'nonleadlike' binders (and bonders) and the proactive selection of those compounds that are 'leadlike' in their binding to the target are vital components of the screening effort. The physiochemical properties of leadlikeness and the surprising differences between those properties and the now classical definitions of druglikeness are becoming apparent.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A280%3ADC%252BD3s%252Fls1SjtQ%253D%253D&md5=5e34cece6f751fa727f949c3526aa95c
39
Rush, T. S., III; Grant, J. A.; Mosyak, L.; Nicholls, A. A shape-based 3-D scaffold hopping method and its application to a bacterial protein-protein interaction J. Med. Chem. 2005, 48, 1489– 1495
[ACS Full Text ], [CAS], Google Scholar
39
A Shape-Based 3-D Scaffold Hopping Method and Its Application to a Bacterial Protein-Protein Interaction
Rush, Thomas S., III; Grant, J. Andrew; Mosyak, Lidia; Nicholls, Anthony
Journal of Medicinal Chemistry (2005), 48 (5), 1489-1495CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
In this paper, the authors describe the first prospective application of the shape-comparison program ROCS (Rapid Overlay of Chem. Structures) to find new scaffolds for small mol. inhibitors of the ZipA-FtsZ protein-protein interaction, a proposed antibacterial target. The shape comparisons are made relative to the crystallog. detd., bioactive conformation of a high-throughput screening (HTS) hit. The use of ROCS led to the identification of a set of novel, weakly binding inhibitors with scaffolds presenting synthetic opportunities to further optimize biol. affinity and lacking development issues assocd. with the HTS lead. These ROCS-identified scaffolds would have been missed using other structural similarity approaches such as ISIS 2D fingerprints. X-ray crystallog. anal. of one of the new inhibitors bound to ZipA reveals that the shape comparison approach very accurately predicted the binding mode. These exptl. results validate this use of ROCS for chemotype switching or "lead hopping" and suggest that it is of general interest for lead identification in drug discovery endeavors.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXht1Ols78%253D&md5=05b6e54a657a3b8c768e63852d871ef6
40
Nicholls, A.; McGaughey, G. B.; Sheridan, R. P.; Good, A. C.; Warren, G.; Mathieu, M.; Muchmore, S. W.; Brown, S. P.; Grant, J. A.; Haigh, J. A.; Nevins, N.; Jain, A. N.; Kelley, B. Molecular shape and medicinal chemistry: a perspective J. Med. Chem. 2010, 53, 3862– 3886
[ACS Full Text ], [CAS], Google Scholar
40
Molecular Shape and Medicinal Chemistry: A Perspective
Nicholls, Anthony; McGaughey, Georgia B.; Sheridan, Robert P.; Good, Andrew C.; Warren, Gregory; Mathieu, Magali; Muchmore, Steven W.; Brown, Scott P.; Grant, J. Andrew; Haigh, James A.; Nevins, Neysa; Jain, Ajay N.; Kelley, Brian
Journal of Medicinal Chemistry (2010), 53 (10), 3862-3886CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
A review article with 111 refs. summarized perspectives of mol. shape and medicinal chem. in drug screening.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhvF2kt7k%253D&md5=85664344e13872527a3dfb2296d34864
41
Sauer, W. H.; Schwarz, M. K. Molecular shape diversity of combinatorial libraries: a prerequisite for broad bioactivity J. Chem. Inf. Comput. Sci. 2003, 43, 987– 1003
[ACS Full Text ], [CAS], Google Scholar
41
Molecular Shape Diversity of Combinatorial Libraries: A Prerequisite for Broad Bioactivity
Sauer, Wolfgang H. B.; Schwarz, Matthias K.
Journal of Chemical Information and Computer Sciences (2003), 43 (3), 987-1003CODEN: JCISD8; ISSN:0095-2338. (American Chemical Society)
A computational method to rapidly assess and visualize the diversity in mol. shape assocd. with a given compd. set has been developed. Normalized ratios of principal moments of inertia are plotted into two-dimensional triangular graphs and then used to compare the shape space covered by different compd. sets, such as combinatorial libraries of varying size and compn. We have further developed a computational method to analyze interset similarity in terms of shape space coverage, which allows the shape redundancy between the different subsets of a given compd. collection to be analyzed in a quant. way. The shape space coverage has been found to originate mainly from the nature and the 3D-geometry (but not the size) of the central scaffold, while the no. and nature of the peripheral substituents and conformational aspects were shown to be of minor importance. Substantial shape space coverage has been correlated with broad biol. activity by applying the same shape anal. to collections of known bioactive compds., such as MDDR and the GOLD-set. The aggregate of our results corroborates the intuitive notion that mol. shape is intimately linked to biol. activity and that a high degree of shape (hence scaffold) diversity in screening collections will increase the odds of addressing a broad range of biol. targets.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXhvF2muro%253D&md5=c07401742457e76b066bed4b43feca1d
42
Lovering, F.; Bikker, J.; Humblet, C. Escape from flatland: increasing saturation as an approach to improving clinical success J. Med. Chem. 2009, 52, 6752– 6756
[ACS Full Text ], [CAS], Google Scholar
42
Escape from Flatland: Increasing Saturation as an Approach to Improving Clinical Success
Lovering, Frank; Bikker, Jack; Humblet, Christine
Journal of Medicinal Chemistry (2009), 52 (21), 6752-6756CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
The medicinal chem. community has become increasingly aware of the value of tracking calcd. phys. properties such as mol. wt., topol. polar surface area, rotatable bonds, and hydrogen bond donors and acceptors. The authors hypothesized that the shift to high-throughput synthetic practices over the past decade may be another factor that may predispose mols. to fail by steering discovery efforts toward achiral, arom. compds. The authors have proposed two simple and interpretable measures of the complexity of mols. prepd. as potential drug candidates. The first is carbon bond satn. as defined by fraction Sp3 (Fsp3) where Fsp3 = (no. of Sp3 hybridized carbons/total carbon count). The second is simply whether a chiral carbon exists in the mol. The authors demonstrate that both complexity (as measured by Fsp3) and the presence of chiral centers correlate with success as compds. transition from discovery, through clin. testing, to drugs. To explain these observations, the authors further demonstrate that satn. correlates with soly., an exptl. phys. property important to success in the drug discovery setting.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXht1KjtLvN&md5=4ca92c30c17c53d77ad376719bad951e
43
Ritchie, T. J.; Macdonald, S. J.; Young, R. J.; Pickett, S. D. The impact of aromatic ring count on compound developability: further insights by examining carbo- and hetero-aromatic and -aliphatic ring types Drug Discovery Today 2011, 16, 164– 171
[Crossref], [PubMed], [CAS], Google Scholar
43
The impact of aromatic ring count on compound developability: further insights by examining carbo- and hetero-aromatic and -aliphatic ring types
Ritchie, Timothy J.; MacDonald, Simon J. F.; Young, Robert J.; Pickett, Stephen D.
Drug Discovery Today (2011), 16 (3/4), 164-171CODEN: DDTOFS; ISSN:1359-6446. (Elsevier B.V.)
A review. The impact of carboarom., heteroarom., carboaliph. and heteroaliph. ring counts and fused arom. ring count on several developability measures (soly., lipophilicity, protein binding, P 450 inhibition and hERG binding) is the topic for this review article. Recent results indicate that increasing ring counts have detrimental effects on developability in the order carboaroms. » heteroaroms. > carboaliphatics > heteroaliphatics, with heteroaliphatics exerting a beneficial effect in many cases. Increasing arom. ring count exerts effects on several developability parameters that are lipophilicity- and size-independent, and fused arom. systems have a beneficial effect relative to their nonfused counterparts. Increasing arom. ring count has a detrimental effect on human bioavailability parameters, and heteroarom. ring count (but not other ring counts) has increased over time in marketed oral drugs.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhvVyqsr0%253D&md5=8e2d0e4e499d2f07d66ca2f385defb2d
44
Clemons, P. A.; Bodycombe, N. E.; Carrinski, H. A.; Wilson, J. A.; Shamji, A. F.; Wagner, B. K.; Koehler, A. N.; Schreiber, S. L. Small molecules of different origins have distinct distributions of structural complexity that correlate with protein-binding profiles Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 18787– 18792
[Crossref], [PubMed], [CAS], Google Scholar
44
Small molecules of different origins have distinct distributions of structural complexity that correlate with protein-binding profiles
Clemons, Paul A.; Bodycombe, Nicole E.; Carrinski, Hyman A.; Wilson, J. Anthony; Shamji, Alykhan F.; Wagner, Bridget K.; Koehler, Angela N.; Schreiber, Stuart L.
Proceedings of the National Academy of Sciences of the United States of America (2010), 107 (44), 18787-18792, S18787/1-S18787/5CODEN: PNASA6; ISSN:0027-8424. (National Academy of Sciences)
Using a diverse collection of small mols. generated from a variety of sources, we measured protein-binding activities of each individual compd. against each of 100 diverse (sequence-unrelated) proteins using small-mol. microarrays. We also analyzed structural features, including complexity, of the small mols. We found that compds. from different sources (com., academic, natural) have different protein-binding behaviors and that these behaviors correlate with general trends in stereochem. and shape descriptors for these compd. collections. Increasing the content of sp3-hybridized and stereogenic atoms relative to compds. from com. sources, which comprise the majority of current screening collections, improved binding selectivity and frequency. The results suggest structural features that synthetic chemists can target when synthesizing screening collections for biol. discovery. Because binding proteins selectively can be a key feature of high-value probes and drugs, synthesizing com- pounds having features identified in this study may result in improved performance of screening collections.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhsVCqsb%252FN&md5=198e298ca5391d7dfd91184680e3c2ec
45
Clemons, P. A.; Wilson, J. A.; Dancik, V.; Muller, S.; Carrinski, H. A.; Wagner, B. K.; Koehler, A. N.; Schreiber, S. L. Quantifying structure and performance diversity for sets of small molecules comprising small-molecule screening collections Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 6817– 6822
[Crossref], [PubMed], [CAS], Google Scholar
45
Quantifying structure and performance diversity for sets of small molecules comprising small-molecule screening collections
Clemons, Paul A.; Wilson, J. Anthony; Dancik, Vlado; Muller, Sandrine; Carrinski, Hyman A.; Wagner, Bridget K.; Koehler, Angela N.; Schreiber, Stuart L.
Proceedings of the National Academy of Sciences of the United States of America (2011), 108 (17), 6817-6822, S6817/1-S6817/8CODEN: PNASA6; ISSN:0027-8424. (National Academy of Sciences)
Using a diverse collection of small mols. we recently found that compd. sets from different sources (com.; academic; natural) have different protein-binding behaviors, and these behaviors correlate with trends in stereochem. complexity for these compd. sets. These results lend insight into structural features that synthetic chemists might target when synthesizing screening collections for biol. discovery. We report extensive characterization of structural properties and diversity of biol. performance for these compds. and expand comparative analyses to include physicochem. properties and three-dimensional shapes of predicted conformers. The results highlight addnl. similarities and differences between the sets, but also the dependence of such comparisons on the choice of mol. descriptors. Using a protein-binding dataset, we introduce an information-theoretic measure to assess diversity of performance with a constraint on specificity. Rather than relying on finding individual active compds., this measure allows rational judgment of compd. subsets as groups. We also apply this measure to publicly available data from ChemBank for the same compd. sets across a diverse group of functional assays. We find that performance diversity of compd. sets is relatively stable across a range of property values as judged by this measure, both in protein-binding studies and functional assays. Because building screening collections with improved performance depends on efficient use of synthetic org. chem. resources, these studies illustrate an important quant. framework to help prioritize choices made in building such collections.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXlslyru7c%253D&md5=d83343549289415060ea55d45118d31f
46
Sadowski, J.; Gasteiger, J. From atoms and bonds to 3-dimensional atomic coordinates - automatic model builders Chem. Rev. 1993, 93, 2567– 2581
[ACS Full Text ], [CAS], Google Scholar
46
From atoms and bonds to three-dimensional atomic coordinates: automatic model builders
Sadowski, Jens; Gasteiger, Johann
Chemical Reviews (Washington, DC, United States) (1993), 93 (7), 2567-81CODEN: CHREAY; ISSN:0009-2665.
A review with ∼75 refs. in which WIZARD, COBRA, CONCORD, and CORINA are discussed.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK3sXmt1GgsLs%253D&md5=fcdb50dfbd981c8f122da06e06e62e27
47
Bemis, G. W.; Murcko, M. A. The properties of known drugs. 1. Molecular frameworks J. Med. Chem. 1996, 39, 2887– 2893
[ACS Full Text ], [CAS], Google Scholar
47
The Properties of Known Drugs. 1. Molecular Frameworks
Bemis, Guy W.; Murcko, Mark A.
Journal of Medicinal Chemistry (1996), 39 (15), 2887-2893CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
To better understand the common features present in drug mols., we use shape description methods to analyze a database of com. available drugs and prep. a list of common drug shapes. A useful way of organizing this structural data is to group the atoms of each drug mol. into ring, linker, framework, and side chain atoms. On the basis of the two-dimensional mol. structures (without regard to atom type, hybridization, and bond order), there are 1179 different frameworks among the 5120 compds. analyzed. However, the shapes of half of the drugs in the database are described by the 32 most frequently occurring frameworks. This suggests that the diversity of shapes in the set of known drugs is extremely low. In our second method of anal., in which atom type, hybridization, and bond order are considered, more diversity is seen; there are 2506 different frameworks among the 5120 compds. in the database, and the most frequently occurring 42 frameworks account for only one-fourth of the drugs. We discuss the possible interpretations of these findings and the way they may be used to guide future drug discovery research.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK28XjvVejtro%253D&md5=5e2c4fdfea9434456a0cca83de4185b3

Cited By

This article is cited by 743 publications.

Adrian Krzyzanowski, Axel Pahl, Michael Grigalunas, Herbert Waldmann. Spacial Score─A Comprehensive Topological Indicator for Small-Molecule Complexity. Journal of Medicinal Chemistry 2023, 66 (18) , 12739-12750. https://doi.org/10.1021/acs.jmedchem.3c00689
Maho Nakata, Toshiyuki Maeda. PubChemQC B3LYP/6-31G*//PM6 Data Set: The Electronic Structures of 86 Million Molecules Using B3LYP/6-31G* Calculations. Journal of Chemical Information and Modeling 2023, 63 (18) , 5734-5754. https://doi.org/10.1021/acs.jcim.3c00899
Gergely Takács, Dávid Havasi, Márk Sándor, Zsolt Dohánics, György T. Balogh, Róbert Kiss. DIY Virtual Chemical Libraries - Novel Starting Points for Drug Discovery. ACS Medicinal Chemistry Letters 2023, 14 (9) , 1188-1197. https://doi.org/10.1021/acsmedchemlett.3c00146
Hyeongwoo Kim, Kyunghoon Lee, Chansu Kim, Jaechang Lim, Woo Youn Kim. DFRscore: Deep Learning-Based Scoring of Synthetic Complexity with Drug-Focused Retrosynthetic Analysis for High-Throughput Virtual Screening. Journal of Chemical Information and Modeling 2023, Article ASAP.
Sibei Guo, Jun Jiang, Hao Ren, Song Wang. Fusion of Multiple Spectra for Investigating Chemical Bonding Properties via Machine Learning. The Journal of Physical Chemistry Letters 2023, 14 (33) , 7461-7468. https://doi.org/10.1021/acs.jpclett.3c01709
Rostislav Fedorov, Ganna Gryn’ova. Unlocking the Potential: Predicting Redox Behavior of Organic Molecules, from Linear Fits to Neural Networks. Journal of Chemical Theory and Computation 2023, 19 (15) , 4796-4814. https://doi.org/10.1021/acs.jctc.3c00355
Yuyang Wang, Changwen Xu, Zijie Li, Amir Barati Farimani. Denoise Pretraining on Nonequilibrium Molecules for Accurate and Transferable Neural Potentials. Journal of Chemical Theory and Computation 2023, 19 (15) , 5077-5087. https://doi.org/10.1021/acs.jctc.3c00289
Jeffrey A. Dewey, Clémence Delalande, Saara-Anne Azizi, Vivian Lu, Dionysios Antonopoulos, Gyorgy Babnigg. Molecular Glue Discovery: Current and Future Approaches. Journal of Medicinal Chemistry 2023, 66 (14) , 9278-9296. https://doi.org/10.1021/acs.jmedchem.3c00449
Guoxiang Zhao, Weiyin Yan, Zirui Wang, Yao Kang, Zuju Ma, Zhi-Gang Gu, Qiao-Hong Li, Jian Zhang. Predict the Polarizability and Order of Magnitude of Second Hyperpolarizability of Molecules by Machine Learning. The Journal of Physical Chemistry A 2023, 127 (29) , 6109-6115. https://doi.org/10.1021/acs.jpca.2c08563
Thorren Kirschbaum, Börries von Seggern, Joachim Dzubiella, Annika Bande, Frank Noé. Machine Learning Frontier Orbital Energies of Nanodiamonds. Journal of Chemical Theory and Computation 2023, 19 (14) , 4461-4473. https://doi.org/10.1021/acs.jctc.2c01275
Sheng-Hsuan Hung, Zong-Rong Ye, Chi-Feng Cheng, Berlin Chen, Ming-Kang Tsai. Enhanced Predictions for the Experimental Photophysical Data Using the Featurized Schnet-Bondstep Approach. Journal of Chemical Theory and Computation 2023, 19 (14) , 4559-4567. https://doi.org/10.1021/acs.jctc.3c00054
Esther Heid, Charles J. McGill, Florence H. Vermeire, William H. Green. Characterizing Uncertainty in Machine Learning for Chemistry. Journal of Chemical Information and Modeling 2023, 63 (13) , 4012-4029. https://doi.org/10.1021/acs.jcim.3c00373
Nutaya Pravalphruekul, Maytus Piriyajitakonkij, Phond Phunchongharn, Supanida Piyayotai. De Novo Design of Molecules with Multiaction Potential from Differential Gene Expression using Variational Autoencoder. Journal of Chemical Information and Modeling 2023, 63 (13) , 3999-4011. https://doi.org/10.1021/acs.jcim.3c00355
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
Dmitrij Rappoport. Statistics and Bias-Free Sampling of Reaction Mechanisms from Reaction Network Models. The Journal of Physical Chemistry A 2023, 127 (24) , 5252-5263. https://doi.org/10.1021/acs.jpca.3c01430
Po-Yu Kao, Ya-Chu Yang, Wei-Yin Chiang, Jen-Yueh Hsiao, Yudong Cao, Alex Aliper, Feng Ren, Alán Aspuru-Guzik, Alex Zhavoronkov, Min-Hsiu Hsieh, Yen-Chu Lin. Exploring the Advantages of Quantum Generative Adversarial Networks in Generative Chemistry. Journal of Chemical Information and Modeling 2023, 63 (11) , 3307-3318. https://doi.org/10.1021/acs.jcim.3c00562
Po-Yen Chen, Kiyou Shibata, Katsumi Hagita, Tomohiro Miyata, Teruyasu Mizoguchi. Prediction of the Ground-State Electronic Structure from Core-Loss Spectra of Organic Molecules by Machine Learning. The Journal of Physical Chemistry Letters 2023, 14 (20) , 4858-4865. https://doi.org/10.1021/acs.jpclett.3c00142
Eric M. Collins, Krishnan Raghavachari. Interpretable Graph-Network-Based Machine Learning Models via Molecular Fragmentation. Journal of Chemical Theory and Computation 2023, 19 (10) , 2804-2810. https://doi.org/10.1021/acs.jctc.2c01308
Lieven Bekaert, Suzuno Akatsuka, Naoto Tanibata, Frank De Proft, Annick Hubin, Mesfin Haile Mamme, Masanobu Nakayama. Assessing the Reactivity of the Na3PS4 Solid-State Electrolyte with the Sodium Metal Negative Electrode Using Total Trajectory Analysis with Neural-Network Potential Molecular Dynamics. The Journal of Physical Chemistry C 2023, 127 (18) , 8503-8514. https://doi.org/10.1021/acs.jpcc.3c02379
Matthew P. Stewart, Scot T. Martin. Machine Learning for Ionization Potentials and Photoionization Cross Sections of Volatile Organic Compounds. ACS Earth and Space Chemistry 2023, 7 (4) , 863-875. https://doi.org/10.1021/acsearthspacechem.3c00009
Carlos Manuel de Armas-Morejón, Luis A. Montero-Cabrera, Angel Rubio, Joaquim Jornet-Somoza. Electronic Descriptors for Supervised Spectroscopic Predictions. Journal of Chemical Theory and Computation 2023, 19 (6) , 1818-1826. https://doi.org/10.1021/acs.jctc.2c01039
Rishi Gurnani, Christopher Kuenneth, Aubrey Toland, Rampi Ramprasad. Polymer Informatics at Scale with Multitask Graph Neural Networks. Chemistry of Materials 2023, 35 (4) , 1560-1567. https://doi.org/10.1021/acs.chemmater.2c02991
Jinzhe Zeng, Yujun Tao, Timothy J. Giese, Darrin M. York. QDπ: A Quantum Deep Potential Interaction Model for Drug Discovery. Journal of Chemical Theory and Computation 2023, 19 (4) , 1261-1275. https://doi.org/10.1021/acs.jctc.2c01172
Song Xia, Dongdong Zhang, Yingkai Zhang. Multitask Deep Ensemble Prediction of Molecular Energetics in Solution: From Quantum Mechanics to Experimental Properties. Journal of Chemical Theory and Computation 2023, 19 (2) , 659-668. https://doi.org/10.1021/acs.jctc.2c01024
Kristian Kříž, Lisa Schmidt, Alfred T. Andersson, Marie-Madeleine Walz, David van der Spoel. An Imbalance in the Force: The Need for Standardized Benchmarks for Molecular Simulation. Journal of Chemical Information and Modeling 2023, 63 (2) , 412-431. https://doi.org/10.1021/acs.jcim.2c01127
Ye Buehler, Jean-Louis Reymond. Molecular Framework Analysis of the Generated Database GDB-13s. Journal of Chemical Information and Modeling 2023, 63 (2) , 484-492. https://doi.org/10.1021/acs.jcim.2c01107
Dongliang Kang, Jun Ma, Ya-Pu Zhao. Perspectives of Machine Learning Development on Kerogen Molecular Model Reconstruction and Shale Oil/Gas Exploitation. Energy & Fuels 2023, 37 (1) , 98-117. https://doi.org/10.1021/acs.energyfuels.2c03307
Megan A. Lim, Song Yang, Huanghao Mai, Alan C. Cheng. Exploring Deep Learning of Quantum Chemical Properties for Absorption, Distribution, Metabolism, and Excretion Predictions. Journal of Chemical Information and Modeling 2022, 62 (24) , 6336-6341. https://doi.org/10.1021/acs.jcim.2c00245
Maximilian Beckers, Nikolas Fechner, Nikolaus Stiefl. 25 Years of Small-Molecule Optimization at Novartis: A Retrospective Analysis of Chemical Series Evolution. Journal of Chemical Information and Modeling 2022, 62 (23) , 6002-6021. https://doi.org/10.1021/acs.jcim.2c00785
Joshua L. Lansford, Brian C. Barnes, Betsy M. Rice, Klavs F. Jensen. Building Chemical Property Models for Energetic Materials from Small Datasets Using a Transfer Learning Approach. Journal of Chemical Information and Modeling 2022, 62 (22) , 5397-5410. https://doi.org/10.1021/acs.jcim.2c00841
Hisham Abdel-Aty, Ian R. Gould. Large-Scale Distributed Training of Transformers for Chemical Fingerprinting. Journal of Chemical Information and Modeling 2022, 62 (20) , 4852-4862. https://doi.org/10.1021/acs.jcim.2c00715
Alex S. Moraes, Gabriel A. Pinheiro, Tuanan C. Lourenço, Mauro C. Lopes, Marcos G. Quiles, Luis G. Dias, Juarez L. F. Da Silva. Screening of the Role of the Chemical Structure in the Electrochemical Stability Window of Ionic Liquids: DFT Calculations Combined with Data Mining. Journal of Chemical Information and Modeling 2022, 62 (19) , 4702-4712. https://doi.org/10.1021/acs.jcim.2c00748
Masato Sumita, Kei Terayama, Ryo Tamura, Koji Tsuda. QCforever: A Quantum Chemistry Wrapper for Everyone to Use in Black-Box Optimization. Journal of Chemical Information and Modeling 2022, 62 (18) , 4427-4434. https://doi.org/10.1021/acs.jcim.2c00812
GuanYa Yang, Wai Yuet Chiu, Jiang Wu, Yi Zhou, ShuGuang Chen, WeiJun Zhou, Jiaqi Fan, GuanHua Chen. Predicting Experimental Heats of Formation via Deep Learning with Limited Experimental Data. The Journal of Physical Chemistry A 2022, 126 (36) , 6295-6300. https://doi.org/10.1021/acs.jpca.2c02957
Min Xie, Xiaonan Sun, Weixing Li, Jiwen Guan, Zhenhao Liang, Yongjun Hu. A Facile Route for the Formation of Complex Nitrogen-Containing Prebiotic Molecules in the Interstellar Medium. The Journal of Physical Chemistry Letters 2022, 13 (34) , 8207-8213. https://doi.org/10.1021/acs.jpclett.2c01857
Florian Spenke, Bernd Hartke. Graph-based Automated Macro-Molecule Assembly. Journal of Chemical Information and Modeling 2022, 62 (16) , 3714-3723. https://doi.org/10.1021/acs.jcim.2c00609
Filipe Menezes, Grzegorz M. Popowicz. ULYSSES: An Efficient and Easy to Use Semiempirical Library for C++. Journal of Chemical Information and Modeling 2022, 62 (16) , 3685-3694. https://doi.org/10.1021/acs.jcim.2c00757
Zong-Rong Ye, Sheng-Hsuan Hung, Berlin Chen, Ming-Kang Tsai. Assessment of Predicting Frontier Orbital Energies for Small Organic Molecules Using Knowledge-Based and Structural Information. ACS Engineering Au 2022, 2 (4) , 360-368. https://doi.org/10.1021/acsengineeringau.2c00011
Emad Al Ibrahim, Aamir Farooq. Transfer Learning Approach to Multitarget Temperature-Dependent Reaction Rate Prediction. The Journal of Physical Chemistry A 2022, 126 (28) , 4617-4629. https://doi.org/10.1021/acs.jpca.2c00713
Jingbai Li, Steven A. Lopez. A Look Inside the Black Box of Machine Learning Photodynamics Simulations. Accounts of Chemical Research 2022, 55 (14) , 1972-1984. https://doi.org/10.1021/acs.accounts.2c00288
Jonathan Shearer, Jose L. Castro, Alastair D. G. Lawson, Malcolm MacCoss, Richard D. Taylor. Rings in Clinical Trials and Drugs: Present and Future. Journal of Medicinal Chemistry 2022, 65 (13) , 8699-8712. https://doi.org/10.1021/acs.jmedchem.2c00473
Kanishka Singh, Jannes Münchmeyer, Leon Weber, Ulf Leser, Annika Bande. Graph Neural Networks for Learning Molecular Excitation Spectra. Journal of Chemical Theory and Computation 2022, 18 (7) , 4408-4417. https://doi.org/10.1021/acs.jctc.2c00255
Huaipan Jiang, Jian Wang, Weilin Cong, Yihe Huang, Morteza Ramezani, Anup Sarma, Nikolay V. Dokholyan, Mehrdad Mahdavi, Mahmut T. Kandemir. Predicting Protein–Ligand Docking Structure with Graph Neural Network. Journal of Chemical Information and Modeling 2022, 62 (12) , 2923-2932. https://doi.org/10.1021/acs.jcim.2c00127
Jiahui Yu, Jike Wang, Hong Zhao, Junbo Gao, Yu Kang, Dongsheng Cao, Zhe Wang, Tingjun Hou. Organic Compound Synthetic Accessibility Prediction Based on the Graph Attention Mechanism. Journal of Chemical Information and Modeling 2022, 62 (12) , 2973-2986. https://doi.org/10.1021/acs.jcim.2c00038
Kiran Sasikumar, Raghavan Ranganathan, Srujan Rokkam, Tapan Desai, Richard Burnes, Peter Cross. Development of Chemical Kinetics Models from Atomistic Reactive Molecular Dynamics Simulations: Application to Iso-octane Combustion and Rubber Ablative Degradation. The Journal of Physical Chemistry A 2022, 126 (21) , 3358-3372. https://doi.org/10.1021/acs.jpca.2c00901
Mohammadamin Tavakoli, Aaron Mood, David Van Vranken, Pierre Baldi. Quantum Mechanics and Machine Learning Synergies: Graph Attention Neural Networks to Predict Chemical Reactivity. Journal of Chemical Information and Modeling 2022, 62 (9) , 2121-2132. https://doi.org/10.1021/acs.jcim.1c01400
Wendy A. Warr, Marc C. Nicklaus, Christos A. Nicolaou, Matthias Rarey. Exploration of Ultralarge Compound Collections for Drug Discovery. Journal of Chemical Information and Modeling 2022, 62 (9) , 2021-2034. https://doi.org/10.1021/acs.jcim.2c00224
Eiichi Kojima, Atsuhiro Iimuro, Mado Nakajima, Hirotaka Kinuta, Naoya Asada, Yusuke Sako, Zenzaburo Nakata, Kentaro Uemura, Shuhei Arita, Shinobu Miki, Chiaki Wakasa-Morimoto, Yuki Tachibana. Pocket-to-Lead: Structure-Based De Novo Design of Novel Non-peptidic HIV-1 Protease Inhibitors Using the Ligand Binding Pocket as a Template. Journal of Medicinal Chemistry 2022, 65 (8) , 6157-6170. https://doi.org/10.1021/acs.jmedchem.1c02217
Ankur Kumar Gupta, Krishnan Raghavachari. Three-Dimensional Convolutional Neural Networks Utilizing Molecular Topological Features for Accurate Atomization Energy Predictions. Journal of Chemical Theory and Computation 2022, 18 (4) , 2132-2143. https://doi.org/10.1021/acs.jctc.1c00504
Yi Hua, Xiaobao Fang, Guomeng Xing, Yuan Xu, Li Liang, Chenglong Deng, Xiaowen Dai, Haichun Liu, Tao Lu, Yanmin Zhang, Yadong Chen. Effective Reaction-Based De Novo Strategy for Kinase Targets: A Case Study on MERTK Inhibitors. Journal of Chemical Information and Modeling 2022, 62 (7) , 1654-1668. https://doi.org/10.1021/acs.jcim.2c00068
Yury Kostyukevich, Sergey Sosnin, Sergey Osipenko, Oxana Kovaleva, Lidiia Rumiantseva, Albert Kireev, Alexander Zherebker, Maxim Fedorov, Evgeny N. Nikolaev. PyFragMS─A Web Tool for the Investigation of the Collision-Induced Fragmentation Pathways. ACS Omega 2022, 7 (11) , 9710-9719. https://doi.org/10.1021/acsomega.1c07272
Penglei Wang, Shuangjia Zheng, Yize Jiang, Chengtao Li, Junhong Liu, Chang Wen, Atanas Patronov, Dahong Qian, Hongming Chen, Yuedong Yang. Structure-Aware Multimodal Deep Learning for Drug–Protein Interaction Prediction. Journal of Chemical Information and Modeling 2022, 62 (5) , 1308-1317. https://doi.org/10.1021/acs.jcim.2c00060
Raimon Fabregat, Alberto Fabrizio, Edgar A. Engel, Benjamin Meyer, Veronika Juraskova, Michele Ceriotti, Clemence Corminboeuf. Local Kernel Regression and Neural Network Approaches to the Conformational Landscapes of Oligopeptides. Journal of Chemical Theory and Computation 2022, 18 (3) , 1467-1479. https://doi.org/10.1021/acs.jctc.1c00813
André F. Oliveira, Juarez L. F. Da Silva, Marcos G. Quiles. Molecular Property Prediction and Molecular Design Using a Supervised Grammar Variational Autoencoder. Journal of Chemical Information and Modeling 2022, 62 (4) , 817-828. https://doi.org/10.1021/acs.jcim.1c01573
Ruocheng Han, Rangsiman Ketkaew, Sandra Luber. A Concise Review on Recent Developments of Machine Learning for the Prediction of Vibrational Spectra. The Journal of Physical Chemistry A 2022, 126 (6) , 801-812. https://doi.org/10.1021/acs.jpca.1c10417
John M. Simmie. C2H5NO Isomers: From Acetamide to 1,2-Oxazetidine and Beyond. The Journal of Physical Chemistry A 2022, 126 (6) , 924-939. https://doi.org/10.1021/acs.jpca.1c09984
Hideo Doi, Kazuaki Z. Takahashi, Takeshi Aoyagi. Screening toward the Development of Fingerprints of Atomic Environments Using Bond-Orientational Order Parameters. ACS Omega 2022, 7 (5) , 4606-4613. https://doi.org/10.1021/acsomega.1c06587
Viktor Zaverkin, Julia Netz, Fabian Zills, Andreas Köhn, Johannes Kästner. Thermally Averaged Magnetic Anisotropy Tensors via Machine Learning Based on Gaussian Moments. Journal of Chemical Theory and Computation 2022, 18 (1) , 1-12. https://doi.org/10.1021/acs.jctc.1c00853
Lei Tao, Vikas Varshney, Ying Li. Benchmarking Machine Learning Models for Polymer Informatics: An Example of Glass Transition Temperature. Journal of Chemical Information and Modeling 2021, 61 (11) , 5395-5413. https://doi.org/10.1021/acs.jcim.1c01031
Tiago Sousa, João Correia, Vítor Pereira, Miguel Rocha. Generative Deep Learning for Targeted Compound Design. Journal of Chemical Information and Modeling 2021, 61 (11) , 5343-5361. https://doi.org/10.1021/acs.jcim.0c01496
Ying Yang, Kun Yao, Matthew P. Repasky, Karl Leswing, Robert Abel, Brian K. Shoichet, Steven V. Jerome. Efficient Exploration of Chemical Space with Docking and Deep Learning. Journal of Chemical Theory and Computation 2021, 17 (11) , 7106-7119. https://doi.org/10.1021/acs.jctc.1c00810
Xiaochu Tong, Xiaohong Liu, Xiaoqin Tan, Xutong Li, Jiaxin Jiang, Zhaoping Xiong, Tingyang Xu, Hualiang Jiang, Nan Qiao, Mingyue Zheng. Generative Models for De Novo Drug Design. Journal of Medicinal Chemistry 2021, 64 (19) , 14011-14027. https://doi.org/10.1021/acs.jmedchem.1c00927
Viktor Zaverkin, David Holzmüller, Ingo Steinwart, Johannes Kästner. Fast and Sample-Efficient Interatomic Neural Network Potentials for Molecules and Materials Based on Gaussian Moments. Journal of Chemical Theory and Computation 2021, 17 (10) , 6658-6670. https://doi.org/10.1021/acs.jctc.1c00527
Philippe Gantzer, Benoit Creton, Carlos Nieto-Draghi. Comparisons of Molecular Structure Generation Methods Based on Fragment Assemblies and Genetic Graphs. Journal of Chemical Information and Modeling 2021, 61 (9) , 4245-4258. https://doi.org/10.1021/acs.jcim.1c00803
Luis Cesar de Azevedo, Gabriel A. Pinheiro, Marcos G. Quiles, Juarez L. F. Da Silva, Ronaldo C. Prati. Systematic Investigation of Error Distribution in Machine Learning Algorithms Applied to the Quantum-Chemistry QM9 Data Set Using the Bias and Variance Decomposition. Journal of Chemical Information and Modeling 2021, 61 (9) , 4210-4223. https://doi.org/10.1021/acs.jcim.1c00503
Ava P. Soleimany, Alexander Amini, Samuel Goldman, Daniela Rus, Sangeeta N. Bhatia, Connor W. Coley. Evidential Deep Learning for Guided Molecular Property Prediction and Discovery. ACS Central Science 2021, 7 (8) , 1356-1367. https://doi.org/10.1021/acscentsci.1c00546
Julia Westermayr, Philipp Marquetand. Machine Learning for Electronically Excited States of Molecules. Chemical Reviews 2021, 121 (16) , 9873-9926. https://doi.org/10.1021/acs.chemrev.0c00749
Oliver T. Unke, Stefan Chmiela, Huziel E. Sauceda, Michael Gastegger, Igor Poltavsky, Kristof T. Schütt, Alexandre Tkatchenko, Klaus-Robert Müller. Machine Learning Force Fields. Chemical Reviews 2021, 121 (16) , 10142-10186. https://doi.org/10.1021/acs.chemrev.0c01111
Aditya Nandy, Chenru Duan, Michael G. Taylor, Fang Liu, Adam H. Steeves, Heather J. Kulik. Computational Discovery of Transition-metal Complexes: From High-throughput Screening to Machine Learning. Chemical Reviews 2021, 121 (16) , 9927-10000. https://doi.org/10.1021/acs.chemrev.1c00347
Bing Huang, O. Anatole von Lilienfeld. Ab Initio Machine Learning in Chemical Compound Space. Chemical Reviews 2021, 121 (16) , 10001-10036. https://doi.org/10.1021/acs.chemrev.0c01303
Michael Tynes, Wenhao Gao, Daniel J. Burrill, Enrique R. Batista, Danny Perez, Ping Yang, Nicholas Lubbers. Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search. Journal of Chemical Information and Modeling 2021, 61 (8) , 3846-3857. https://doi.org/10.1021/acs.jcim.1c00670
Luis Itza Vazquez-Salazar, Eric D. Boittier, Oliver T. Unke, Markus Meuwly. Impact of the Characteristics of Quantum Chemical Databases on Machine Learning Prediction of Tautomerization Energies. Journal of Chemical Theory and Computation 2021, 17 (8) , 4769-4785. https://doi.org/10.1021/acs.jctc.1c00363
Shachar Fite, Omri Nitecki, Zeev Gross. Custom Tokenization Dictionary, CUSTODI: A General, Fast, and Reversible Data-Driven Representation and Regressor. Journal of Chemical Information and Modeling 2021, 61 (7) , 3285-3291. https://doi.org/10.1021/acs.jcim.1c00563
Logan Ward, Naveen Dandu, Ben Blaiszik, Badri Narayanan, Rajeev S. Assary, Paul C. Redfern, Ian Foster, Larry A. Curtiss. Graph-Based Approaches for Predicting Solvation Energy in Multiple Solvents: Open Datasets and Machine Learning Models. The Journal of Physical Chemistry A 2021, 125 (27) , 5990-5998. https://doi.org/10.1021/acs.jpca.1c01960
SahaIshikaGraduate Student Researcherishikasaha@g.ucla.eduHarranPatrick G.D.J. & J.M. Cram Chair in Organic Chemistryharran@chem.ucla.eduDr. Jonathan Bohmann, Department of Pharmaceuticals and Bioengineering, Southwest Research Institute, Ryan Gumpper, Postdoctoral Researcher, University of North Carolina at Chapel Hill. Virtual Screening for Chemists. 2021https://doi.org/10.1021/acsinfocus.7e5001
Alan E. Bilsland, Kirsten McAulay, Ryan West, Angelo Pugliese, Justin Bower. Automated Generation of Novel Fragments Using Screening Data, a Dual SMILES Autoencoder, Transfer Learning and Syntax Correction. Journal of Chemical Information and Modeling 2021, 61 (6) , 2547-2559. https://doi.org/10.1021/acs.jcim.0c01226
Maarten R. Dobbelaere, Pieter P. Plehiers, Ruben Van de Vijver, Christian V. Stevens, Kevin M. Van Geem. Learning Molecular Representations for Thermochemistry Prediction of Cyclic Hydrocarbons and Oxygenates. The Journal of Physical Chemistry A 2021, 125 (23) , 5166-5179. https://doi.org/10.1021/acs.jpca.1c01956
Guo-Li Xiong, Yue Zhao, Lu Liu, Zhong-Ye Ma, Ai-Ping Lu, Yan Cheng, Ting-Jun Hou, Dong-Sheng Cao. Computational Bioactivity Fingerprint Similarities To Navigate the Discovery of Novel Scaffolds. Journal of Medicinal Chemistry 2021, 64 (11) , 7544-7554. https://doi.org/10.1021/acs.jmedchem.1c00234
R. Han, S. Luber. Fast Estimation of Møller–Plesset Correlation Energies Based on Atomic Contributions. The Journal of Physical Chemistry Letters 2021, 12 (22) , 5324-5331. https://doi.org/10.1021/acs.jpclett.1c00900
Ricardo M. Borges, Sean M. Colby, Susanta Das, Arthur S. Edison, Oliver Fiehn, Tobias Kind, Jesi Lee, Amy T. Merrill, Kenneth M. Merz, Jr., Thomas O. Metz, Jamie R. Nunez, Dean J. Tantillo, Lee-Ping Wang, Shunyang Wang, Ryan S. Renslow. Quantum Chemistry Calculations for Metabolomics. Chemical Reviews 2021, 121 (10) , 5633-5670. https://doi.org/10.1021/acs.chemrev.0c00901
Felix Mayr, Alessio Gagliardi. Global Property Prediction: A Benchmark Study on Open-Source, Perovskite-like Datasets. ACS Omega 2021, 6 (19) , 12722-12732. https://doi.org/10.1021/acsomega.1c00991
Pingshi Yu, Alistair J. Sterling, Jotun Hein. A Novel Automated Screening Method for Combinatorially Generated Small Molecules. Journal of Chemical Information and Modeling 2021, 61 (4) , 1637-1646. https://doi.org/10.1021/acs.jcim.0c01462
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
Juliette Zito, Ivan Infante. The Future of Ligand Engineering in Colloidal Semiconductor Nanocrystals. Accounts of Chemical Research 2021, 54 (7) , 1555-1564. https://doi.org/10.1021/acs.accounts.0c00765
Jianing Lu, Song Xia, Jieyu Lu, Yingkai Zhang. Dataset Construction to Explore Chemical Space with 3D Geometry and Deep Learning. Journal of Chemical Information and Modeling 2021, 61 (3) , 1095-1104. https://doi.org/10.1021/acs.jcim.1c00007
Dakota L. Folmsbee, David R. Koes, Geoffrey R. Hutchison. Evaluation of Thermochemical Machine Learning for Potential Energy Curves and Geometry Optimization. The Journal of Physical Chemistry A 2021, 125 (9) , 1987-1993. https://doi.org/10.1021/acs.jpca.0c10147
Felicity F. Nielson, Sean M. Colby, Dennis G. Thomas, Ryan S. Renslow, Thomas O. Metz. Exploring the Impacts of Conformer Selection Methods on Ion Mobility Collision Cross Section Predictions. Analytical Chemistry 2021, 93 (8) , 3830-3838. https://doi.org/10.1021/acs.analchem.0c04341
Pablo A. Unzueta, Chandler S. Greenwell, Gregory J. O. Beran. Predicting Density Functional Theory-Quality Nuclear Magnetic Resonance Chemical Shifts via Δ-Machine Learning. Journal of Chemical Theory and Computation 2021, 17 (2) , 826-840. https://doi.org/10.1021/acs.jctc.0c00979
Jon Paul Janet, Chenru Duan, Aditya Nandy, Fang Liu, Heather J. Kulik. Navigating Transition-Metal Chemical Space: Artificial Intelligence for First-Principles Design. Accounts of Chemical Research 2021, 54 (3) , 532-545. https://doi.org/10.1021/acs.accounts.0c00686
Yao Shi, Paloma L. Prieto, Tara Zepel, Shad Grunert, Jason E. Hein. Automated Experimentation Powers Data Science in Chemistry. Accounts of Chemical Research 2021, 54 (3) , 546-555. https://doi.org/10.1021/acs.accounts.0c00736
Joydeep Munshi, Wei Chen, TeYu Chien, Ganesh Balasubramanian. Transfer Learned Designer Polymers For Organic Solar Cells. Journal of Chemical Information and Modeling 2021, 61 (1) , 134-142. https://doi.org/10.1021/acs.jcim.0c01157
Wenhao Gao, Connor W. Coley. The Synthesizability of Molecules Proposed by Generative Models. Journal of Chemical Information and Modeling 2020, 60 (12) , 5714-5723. https://doi.org/10.1021/acs.jcim.0c00174
Obaidur Rahaman, Alessio Gagliardi. Deep Learning Total Energies and Orbital Energies of Large Organic Molecules Using Hybridization of Molecular Fingerprints. Journal of Chemical Information and Modeling 2020, 60 (12) , 5971-5983. https://doi.org/10.1021/acs.jcim.0c00687
Beomchang Kang, Chaok Seok, Juyong Lee. Prediction of Molecular Electronic Transitions Using Random Forests. Journal of Chemical Information and Modeling 2020, 60 (12) , 5984-5994. https://doi.org/10.1021/acs.jcim.0c00698
Maho Nakata, Tomomi Shimazaki, Masatomo Hashimoto, Toshiyuki Maeda. PubChemQC PM6: Data Sets of 221 Million Molecules with Optimized Molecular Geometries and Electronic Properties. Journal of Chemical Information and Modeling 2020, 60 (12) , 5891-5899. https://doi.org/10.1021/acs.jcim.0c00740
David Balcells, Bastian Bjerkem Skjelstad. tmQM Dataset—Quantum Geometries and Properties of 86k Transition Metal Complexes. Journal of Chemical Information and Modeling 2020, 60 (12) , 6135-6146. https://doi.org/10.1021/acs.jcim.0c01041
Sebastian Mosbach, Angiras Menon, Feroz Farazi, Nenad Krdzavac, Xiaochi Zhou, Jethro Akroyd, Markus Kraft. Multiscale Cross-Domain Thermochemical Knowledge-Graph. Journal of Chemical Information and Modeling 2020, 60 (12) , 6155-6166. https://doi.org/10.1021/acs.jcim.0c01145
Marina P. Oliveira, Maurice Andrey, Salomé R. Rieder, Leyla Kern, David F. Hahn, Sereina Riniker, Bruno A. C. Horta, Philippe H. Hünenberger. Systematic Optimization of a Fragment-Based Force Field against Experimental Pure-Liquid Properties Considering Large Compound Families: Application to Saturated Haloalkanes. Journal of Chemical Theory and Computation 2020, 16 (12) , 7525-7555. https://doi.org/10.1021/acs.jctc.0c00683
Narasimharao Mukku, Prabhakara Madivalappa Davanagere, Kaushik Chanda, Barnali Maiti. A Facile Microwave-Assisted Synthesis of Oxazoles and Diastereoselective Oxazolines Using Aryl-Aldehydes, p-Toluenesulfonylmethyl Isocyanide under Controlled Basic Conditions. ACS Omega 2020, 5 (43) , 28239-28248. https://doi.org/10.1021/acsomega.0c04130
B. Christopher Rinderspacher. Heuristic Global Optimization in Chemical Compound Space. The Journal of Physical Chemistry A 2020, 124 (43) , 9044-9060. https://doi.org/10.1021/acs.jpca.0c05941

Load more citations

Download PDF

Abstract
High Resolution Image
Download MS PowerPoint Slide
Figure 1
Figure 1. Enumeration of GDB-17 starting from mathematical graphs.
High Resolution Image
Download MS PowerPoint Slide
Figure 2
Figure 2. Size and MW profiles of the enumerated chemical space in GDB and the reference databases PubChem, ChEMBL, and DrugBank. The size of the leadlike subsets of GDB (GDBLL, GDBLLnoSR) is extrapolated from analyzing a 1% random subset of GDB-17.
High Resolution Image
Download MS PowerPoint Slide
Figure 3
Figure 3. Drugs and examples of isomers found in GDB-17. All isomers shown have a shape similarity score ROCS > 1.4. None of the isomers shown are known (Scifinder search). Only acyclovir does not occur in GDB-17 because it contains a hemiaminal (N–Csp³–O), a functional group which is excluded from the enumeration.
High Resolution Image
Download MS PowerPoint Slide
Figure 4
Figure 4. Molecule topologies and categories in GDB-17 and reference databases. A. Percentage of reference database compatible with GDB-17 enumeration rules or excluded due to nonenumerated halogen (acyl halide, aliphatic halocarbons) or sulfur (thiols, thioethers), functional groups (acyclic acetals, hemiacetals, aminals, azides, aliphatic nitro groups), element (P, Si, B, Bi, Hg, etc.), skeleton (nonaromatic C═C), or graph (e.g., small rings at 17 atoms). B. Fraction of compounds with small rings. C. Topologies D. Database contents as function of molecular categories. Molecules are assigned to one category only with priority order heteroaromatic > aromatic > heterocyclic > carbocyclic > acyclic. The data for GDB-17 and its subsets were computed from a 1% random subset of the database.
High Resolution Image
Download MS PowerPoint Slide
Figure 5
Figure 5. Polarity features. A. c logP histogram in intervals −5.5 to −4.5, −4.5 to −3.5, etc; B. Average clogP as function of hac; C. H-bond donor atom (HBD) histogram; D. Average HBD as function of hac. The data for GDB-17 and its subsets were computed from a 1% random subset of the database.
High Resolution Image
Download MS PowerPoint Slide
Figure 6
Figure 6. Molecular shape analyzed by the principal moments of inertia. (41) Occupancy maps are shown in the (P1,P2)-plane, in which P1 and P2 are the normalized ratios of the principal moments of inertia (for details see section Methods), and are colored from blue (1 cpd/pixel) to purple (maximum cpd/pixel for each map: GDB-17: 4,691, GDBLL-17: 889, GDBLLnoSR-17: 684, Pubchem-17: 6202, Chembl-17: 487, Drugbank-17: 4). The inserts show an enlarged view of the lower left edge of each triangle where occupancy is highest for PubChem-17, ChEMBL-17, and DrugBank-17. The GDB-17, GDBLL-17, and GDBLLnoSR-17 were analyzed with a random subset of 16.7 million molecules from GDB-17. For all compounds a single stereoisomer was analyzed as generated by CORINA.
High Resolution Image
Download MS PowerPoint Slide
Figure 7
Figure 7. Histograms of quaternary centers (qv) and bonds in fused rings (bfr) in the different databases. The data for GDB-17 and its subsets were computed from a 1% random subset of the database.
High Resolution Image
Download MS PowerPoint Slide
Figure 8
Figure 8. Stereochemistry. A Numbers of stereoisomers per compounds. B. Average number of stereoisomer per compound as a function of hac. Stereoisomers were generated from SMILES using CORINA. The data for GDB-17, GDBLL-17, and GDBLLnoSR-17 stem from the analysis of a random 16.7 million subset of GDB-17.
High Resolution Image
Download MS PowerPoint Slide
Figure 9
Figure 9. Examples of yet unknown C₁₇-ring systems from GDB-17. These hydrocarbons do not give any hits in Scifinder using ″any atom″ types for carbons and ″any bond″ for bonds, including substructure searches but locking further ring fusions. Stereochemistry is not considered in these searches. The ring systems are shown as one possible stereoisomer.
High Resolution Image
Download MS PowerPoint Slide
References
ARTICLE SECTIONS
Jump To
This article references 47 other publications.
1. 1
  Lipkus, A. H.; Yuan, Q.; Lucas, K. A.; Funk, S. A.; Bartelt, W. F.; Schenck, R. J.; Trippe, A. J. Structural diversity of organic chemistry. A scaffold analysis of the CAS Registry J. Org. Chem. 2008, 73, 4443– 4451
  [ACS Full Text ], [CAS], Google Scholar
  1
  Structural Diversity of Organic Chemistry. A Scaffold Analysis of the CAS Registry
  Lipkus, Alan H.; Yuan, Qiong; Lucas, Karen A.; Funk, Susan A.; Bartelt, William F., III; Schenck, Roger J.; Trippe, Anthony J.
  Journal of Organic Chemistry (2008), 73 (12), 4443-4451CODEN: JOCEAH; ISSN:0022-3263. (American Chemical Society)
  The anal. of chem. diversity has become a topic of considerable interest in recent years. This interest has been stimulated by the challenge of discovering novel small-mol. pharmaceuticals. The development of technologies such as combinatorial synthesis and high-throughput screening has made it possible to explore drug-like regions of chem. space in relatively short times. Chem. space is vast and the problem of selecting which region of that space to explore remains a key issue in drug discovery. By analyzing the scaffold content of the CAS Registry, the authors attempt to characterize in a comprehensive way the structural diversity of org. chem. The scaffold of a mol. is taken to be its framework, defined as all its ring systems and all the linkers that connect them. Framework data from more than 24 million org. compds. are analyzed. The distribution of frameworks among compds. is found to be top-heavy, i.e., a small percentage of frameworks occurs in a large percentage of compds. When frameworks are analyzed at the graph level, an even more top-heavy distribution is found: half of the compds. can be described by only 143 framework shapes. The most significant finding is that the framework distribution conforms almost exactly to a power law. This suggests that the more often a framework has been used as the basis for a compd., the more likely it is to be used in another compd. This may be explained by the cost of synthesis: making a new deriv. of a framework is probably less costly if many other derivs. are known. The authors believe that this power law is evidence that the minimization of synthetic cost has been a key factor in shaping the known universe of org. chem.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXmtlylu7c%253D&md5=c03682d3a3fab0b0eec7a36ef1ecc778
2. 2
  ACS NEWS Chem. Eng. News 2011, 89, 38
  Google Scholar
  There is no corresponding record for this reference.
3. 3
  Bleicher, K. H.; Bohm, H. J.; Muller, K.; Alanine, A. I. Hit and lead generation: Beyond high-throughput screening Nat. Rev. Drug Discovery 2003, 2, 369– 378
  [Crossref], [PubMed], [CAS], Google Scholar
  3
  A guide to drug discovery: Hit and lead generation: beyond high-throughput screening
  Bleicher, Konrad H.; Boehm, Hans-Joachim; Mueller, Klaus; Alanine, Alexander I.
  Nature Reviews Drug Discovery (2003), 2 (5), 369-378CODEN: NRDDAG; ISSN:1474-1776. (Nature Publishing Group)
  A review. The identification of small-mol. modulators of protein function, and the process of transforming these into high-content lead series, are key activities in modern drug discovery. The decisions taken during this process have far-reaching consequences for success later in lead optimization and even more crucially in clin. development. Recently, there has been an increased focus on these activities due to escalating downstream costs resulting from high clin. failure rates. In addn., the vast emerging opportunities from efforts in functional genomics and proteomics demands a departure from the linear process of identification, evaluation and refinement activities towards a more integrated parallel process. This calls for flexible, fast and cost-effective strategies to meet the demands of producing high-content lead series with improved prospects for clin. success.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXjslamtb8%253D&md5=b54e55c1f0d34fbd172c8aa9b996591d
4. 4
  Schreiber, S. L. Small molecules: the missing link in the central dogma Nat. Chem. Biol. 2005, 1, 64– 66
  [Crossref], [PubMed], [CAS], Google Scholar
  4
  Small molecules: the missing link in the central dogma
  Schreiber, Stuart L.
  Nature Chemical Biology (2005), 1 (2), 64-66CODEN: NCBABT; ISSN:1552-4450. (Nature Publishing Group)
  There is no expanded citation for this reference.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXls1Oms7Y%253D&md5=d4a54994c0d933602d9d02747e5e1ddc
5. 5
  Mayr, L. M.; Bojanic, D. Novel trends in high-throughput screening Curr. Opin. Pharmacol. 2009, 9, 580– 588
  [Crossref], [PubMed], [CAS], Google Scholar
  5
  Novel trends in high-throughput screening
  Mayr, Lorenz M.; Bojanic, Dejan
  Current Opinion in Pharmacology (2009), 9 (5), 580-588CODEN: COPUBK; ISSN:1471-4892. (Elsevier B.V.)
  A review. High-throughput screening (HTS) is a well-established process for lead discovery in Pharma and Biotech companies and is now also being used for basic and applied research in academia. It comprises the screening of large chem. libraries for activity against biol. targets via the use of automation, miniaturized assays and large-scale data anal. Since its first advent in the early to mid 1990s, the field of HTS has seen not only a continuous change in technol. and processes, but also an adaptation to various needs in lead discovery. HTS has now evolved into a mature discipline that is a crucial source of chem. starting points for drug discovery. Whereas in previous years much emphasis has been put on a steady increase in screening capacity (quant. increase') via automation and miniaturization, the past years have seen a much greater emphasis on content and quality (qual. increase'). Today, many experts in the field see HTS at a crossroad with the need to decide on either higher throughput/more experimentation or a greater focus on assays of greater physiol. relevance, both of which may lead to higher productivity in pharmaceutical R&D. In this paper, we describe the development of HTS over the past decade and point out our own ideas for future directions of HTS in biomedical research. We predict that the trend toward further miniaturization will slow down with the balanced implementation of 384 well, 1536 well, and 384 low vol. well plates. Furthermore, we envisage that there will be much more emphasis on rigorous assay and chem. characterization, particularly considering that novel and more difficult target classes will be pursued. In recent years we have witnessed a clear trend in the drug discovery community toward rigorous hit validation by the use of orthogonal readout technologies, label free and biophys. methodologies. We also see a trend toward a more flexible use of the various screening approaches in lead discovery, i.e., the use of both full deck compd. screening as well as the use of focused screening and iterative screening approaches. Moreover, we expect greater usage of target identification strategies downstream of phenotypic screening and the more effective implementation of affinity selection technologies as a result of advances in chem. diversity methodologies. We predict that, ultimately, each hit finding strategy will be much more project-related, tailor-made, and better integrated into the broader drug discovery efforts.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXht1WntLnL&md5=50284fdd9f4d610e81fb442b78b7b69e
6. 6
  Renner, S.; Popov, M.; Schuffenhauer, A.; Roth, H. J.; Breitenstein, W.; Marzinzik, A.; Lewis, I.; Krastel, P.; Nigsch, F.; Jenkins, J.; Jacoby, E. Recent trends and observations in the design of high-quality screening collections Future Med. Chem 2011, 3, 751– 766
  [Crossref], [PubMed], [CAS], Google Scholar
  6
  Recent trends and observations in the design of high-quality screening collections
  Renner, Steffen; Popov, Maxim; Schuffenhauer, Ansgar; Roth, Hans-Joerg; Breitenstein, Werner; Marzinzik, Andreas; Lewis, Ian; Krastel, Philipp; Nigsch, Florian; Jenkins, Jeremy; Jacoby, Edgar
  Future Medicinal Chemistry (2011), 3 (6), 751-766CODEN: FMCUA7; ISSN:1756-8919. (Future Science Ltd.)
  A review. The design of a high-quality screening collection is of utmost importance for the early drug-discovery process and provides, in combination with high-quality assay systems, the foundation of future discoveries. Herein, we review recent trends and observations to successfully expand the access to bioactive chem. space, including the feedback from hit assessment interviews of high-throughput screening campaigns; recent successes with chemogenomics target family approaches, the identification of new relevant target/domain families, diversity-oriented synthesis and new emerging compd. classes, and non-classical approaches, such as fragment-based screening and DNA-encoded chem. libraries. The role of in silico library design approaches are emphasized.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXlvFGisLY%253D&md5=f876f165f937c9792c989053492e59f5
7. 7
  Kola, I.; Landis, J. Can the pharmaceutical industry reduce attrition rates? Nat. Rev. Drug Discovery 2004, 3, 711– 715
  [Crossref], [PubMed], [CAS], Google Scholar
  7
  Opinion: Can the pharmaceutical industry reduce attrition rates?
  Kola, Ismail; Landis, John
  Nature Reviews Drug Discovery (2004), 3 (8), 711-716CODEN: NRDDAG; ISSN:1474-1776. (Nature Publishing Group)
  The pharmaceutical industry faces considerable challenges, both politically and fiscally. Politically, governments around the world are trying to contain costs and, as health care budgets constitute a very significant part of governmental spending, these costs are the subject of intense scrutiny. In the United States, drug costs are also the subject of intense political discourse. This article deals with the fiscal pressures that face the industry from the perspective of R&D. What impinges on productivity How can we improve current reduced R&D productivity.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXmtVOhtLs%253D&md5=f9025c13a1506f607aaf68415570ed01
8. 8
  Hann, M. M. Molecular obesity, potency and other addictions in drug discovery MedChemComm 2011, 2, 349– 355
  [Crossref], [CAS], Google Scholar
  8
  Molecular obesity, potency and other addictions in drug discovery
  Hann, Michael M.
  MedChemComm (2011), 2 (5), 349-355CODEN: MCCEAY; ISSN:2040-2503. (Royal Society of Chemistry)
  A review. Despite the increase in global biol. and chem. knowledge the discovery of effective and safe new drugs seems to become harder rather than easier. Some of this challenge is due to increasing demands for safety and novelty, but some of the risk involved in this should be controllable if we had more effectively learned from our failures. This perspective reflects on some of the learnings of recent years in relation to the causes of attrition. The term Mol. Obesity is introduced to describe our tendency to build potency into mols. by the inappropriate use of lipophilicity which leads to the premature demise of drug candidates.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXls1Kht7g%253D&md5=954870db4b1d9e846612c3bb256582f7
9. 9
  Schneider, G.; Fechner, U. Computer-based de novo design of drug-like molecules Nat. Rev. Drug Discovery 2005, 4, 649– 663
  [Crossref], [PubMed], [CAS], Google Scholar
  9
  Computer-based de novo design of drug-like molecules
  Schneider, Gisbert; Fechner, Uli
  Nature Reviews Drug Discovery (2005), 4 (8), 649-663CODEN: NRDDAG; ISSN:1474-1776. (Nature Publishing Group)
  A review with refs. Ever since the first automated de novo design techniques were conceived only 15 years ago, the computer-based design of hit and lead structure candidates has emerged as a complementary approach to high-throughput screening. Although many challenges remain, de novo design supports drug discovery projects by generating novel pharmaceutically active agents with desired properties in a cost- and time-efficient manner. In this review, we outline the various design concepts and highlight current developments in computer-based de novo design.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXmvVOqtro%253D&md5=a30dbc58ed81e0b7fe3f7d41a668e9ac
10. 10
  Jorgensen, W. L. Efficient drug lead discovery and optimization Acc. Chem. Res. 2009, 42, 724– 733
  [ACS Full Text ], [CAS], Google Scholar
  10
  Efficient Drug Lead Discovery and Optimization
  Jorgensen, William L.
  Accounts of Chemical Research (2009), 42 (6), 724-733CODEN: ACHRE4; ISSN:0001-4842. (American Chemical Society)
  A review. During the 1980s, advances in the abilities to perform computer simulations of chem. and biomol. systems and to calc. free energy changes led to the expectation that such methodol. would soon show great utility for guiding mol. design. Important potential applications included design of selective receptors, catalysts, and regulators of biol. function including enzyme inhibitors. This time also saw the rise of high-throughput screening and combinatorial chem. along with complementary computational methods for de novo design and virtual screening including docking. These technologies appeared poised to deliver diverse lead compds. for any biol. target. As with many technol. advances, realization of the expectations required significant addnl. effort and time. However, as summarized here, striking success has now been achieved for computer-aided drug lead generation and optimization. De novo design using both mol. growing and docking are illustrated for lead generation, and lead optimization features free energy perturbation calcns. in conjunction with Monte Carlo statistical mechanics simulations for protein-inhibitor complexes in aq. soln. The specific applications are to the discovery of non-nucleoside inhibitors of HIV reverse transcriptase (HIV-RT) and inhibitors of the binding of the proinflammatory cytokine MIF to its receptor CD74. A std. protocol is presented that includes scans for possible addns. of small substituents to a mol. core, interchange of heterocycles, and focused optimization of substituents at one site. Initial leads with activities at low-micromolar concns. have been advanced rapidly to low-nanomolar inhibitors.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXjsFentr4%253D&md5=9641587769701c6541b38e18fa05538a
11. 11
  Reymond, J. L.; Van Deursen, R.; Blum, L. C.; Ruddigkeit, L. Chemical space as a source for new drugs MedChemComm 2010, 1, 30– 38
  [Crossref], [CAS], Google Scholar
  11
  Chemical space as a source for new drugs
  Reymond, Jean-Louis; van Deursen, Ruud; Blum, Lorenz C.; Ruddigkeit, Lars
  MedChemComm (2010), 1 (1), 30-38CODEN: MCCEAY; ISSN:2040-2503. (Royal Society of Chemistry)
  A review. The chem. space is the ensemble of all possible mols., which is believed to contain at least 1060 org. mols. below 500 Da of possible interest for drug discovery. This review summarizes the development of the chem. space concept from enumerating acyclic hydrocarbons in the 1800's to the recent assembly of the chem. universe database GDB. Chem. space travel algorithms can be used to explore defined regions of chem. space by generating focused virtual libraries. Maps of the chem. space are produced from property spaces visualized by principal component anal. or by self-organizing maps, and from structural analyses such as the scaffold-tree or the MQN-system. Virtual screening of virtual chem. space followed by synthesis and testing of the best hits leads to the discovery of new drug mols.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhtVeju7rF&md5=579428edc79f706101ab58c505052c72
12. 12
  Hartenfeller, M.; Schneider, G. De novo drug design Methods Mol. Biol. 2011, 672, 299– 323
  [Crossref], [PubMed], [CAS], Google Scholar
  12
  De novo drug design
  Hartenfeller, Markus; Schneider, Gisbert
  Methods in Molecular Biology (New York, NY, United States) (2011), 672 (Chemoinformatics and Computational Chemical Biology), 299-323CODEN: MMBIED; ISSN:1064-3745. (Springer)
  A review. Computer-assisted mol. design supports drug discovery by suggesting novel chemotypes and compd. modifications for lead structure optimization. While the aspect of synthetic feasibility of the automatically designed compds. has been neglected for a long time, we are currently witnessing an increased interest in this topic. Here, we review state-of-the-art software for de novo drug design with a special emphasis on fragment-based techniques that generate druglike, synthetically accessible compds. The importance of scoring functions that can be used to predict compd. reactivity and potency is highlighted, and several promising solns. are discussed. Recent practical validation studies are presented that have already demonstrated that rule-based fragment assembly can result in novel synthesizable compds. with druglike properties and a desired biol. activity.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhtlejsr7L&md5=3a30147da9187fa4895ba807cefed82a
13. 13
  Klebe, G. Virtual ligand screening: strategies, perspectives and limitations Drug Discovery Today 2006, 11, 580– 594
  [Crossref], [PubMed], [CAS], Google Scholar
  13
  Virtual ligand screening: strategies, perspectives and limitations
  Klebe, Gerhard
  Drug Discovery Today (2006), 11 (13 & 14), 580-594CODEN: DDTOFS; ISSN:1359-6446. (Elsevier B.V.)
  A review. In contrast to high-throughput screening, in virtual ligand screening (VS), compds. are selected using computer programs to predict their binding to a target receptor. A key prerequisite is knowledge about the spatial and energetic criteria responsible for protein-ligand binding. The concepts and prerequisites to perform VS are summarized here, and explanations are sought for the enduring limitations of the technol. Target selection, anal. and prepn. are discussed, as well as considerations about the compilation of candidate ligand libraries. The tools and strategies of a VS campaign, and the accuracy of scoring and ranking of the results, are also considered.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD28XlvFGqtLo%253D&md5=5adfd125d48082238a6c5ad0e8343c59
14. 14
  Kolb, P.; Ferreira, R. S.; Irwin, J. J.; Shoichet, B. K. Docking and chemoinformatic screens for new ligands and targets Curr. Opin. Biotechnol. 2009, 20, 429– 36
  [Crossref], [PubMed], [CAS], Google Scholar
  14
  Docking and chemoinformatic screens for new ligands and targets
  Kolb, Peter; Ferreira, Rafaela S.; Irwin, John J.; Shoichet, Brian K.
  Current Opinion in Biotechnology (2009), 20 (4), 429-436CODEN: CUOBE3; ISSN:0958-1669. (Elsevier B.V.)
  A review. Computer-based docking screens are now widely used to discover new ligands for targets of known structure; in the last two years alone, the discovery of ligands for more than 20 proteins has been reported. Recently, investigators have also turned to predicting new substrates for enzymes of unknown function, taking docking in a wholly new direction. Increasingly, the hit rates, the true-positives, and the false-positives from the docking screens are being compared to those from empirical, high-throughput screens, revealing the strengths, weaknesses, and complementarities of both techniques. The recent efflorescence of GPCR structures has made these quintessential drug targets available to structure-based approaches. Consistent with their druggability', the docking screens have returned high hit rates and potent mols. Finally, in the last several years, an approach almost exactly opposite to docking has also appeared; this pharmacol. network approach begins not with the structure of the target but rather those of drug mols. and asks, given a pattern of chem. in the ligands, what targets may a particular drug bind to. This method, which returns to an older, pharmacol. logic, has been surprisingly successful in predicting new off-targets' for established drugs.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXht1OrtLbL&md5=4b6b2457a735e18be5780d9299ebffe6
15. 15
  Geppert, H.; Vogt, M.; Bajorath, J. Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation J. Chem. Inf. Model. 2010, 50, 205– 216
  [ACS Full Text ], [CAS], Google Scholar
  15
  Current Trends in Ligand-Based Virtual Screening: Molecular Representations, Data Mining Methods, New Application Areas, and Performance Evaluation
  Geppert, Hanna; Vogt, Martin; Bajorath, Jurgen
  Journal of Chemical Information and Modeling (2010), 50 (2), 205-216CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  There is no expanded citation for this reference.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXnvFWmsA%253D%253D&md5=aeb674ef7c6c93711eb21452732e21d8
16. 16
  Cayley, E. Ueber die analytischen Figuren, welche in der Mathematik Bäume genannt werden und ihre Anwendung auf die Theorie chemischer Verbindungen Chem. Ber. 1875, 8, 1056– 1059
  [Crossref], Google Scholar
  There is no corresponding record for this reference.
17. 17
  Lederberg, J.; Sutherland, G. L.; Buchanan, B. G.; Feigenbaum, E. A.; Robertson, A. V.; Duffield, A. M.; Djerassi, C. Applications of artificial intelligence for chemical inference. I. Number of possible organic compounds. Acyclic structures containing carbon, hydrogen, oxygen, and nitrogen J. Am. Chem. Soc. 1969, 91, 2973– 2976
  [ACS Full Text ], Google Scholar
  There is no corresponding record for this reference.
18. 18
  Steinbeck, C. Recent developments in automated structure elucidation of natural products Nat. Prod. Rep. 2004, 21, 512– 518
  [Crossref], [PubMed], [CAS], Google Scholar
  18
  Recent developments in automated structure elucidation of natural products
  Steinbeck, Christoph
  Natural Product Reports (2004), 21 (4), 512-518CODEN: NPRRDF; ISSN:0265-0568. (Royal Society of Chemistry)
  A review. Advancements in the field of computer-assisted structure elucidation (CASE) of natural products achieved in the past five years are discussed. This process starts with a dereplication procedure, supported by structure-spectrum databases. Both com. and free products are available to support the procedure. A no. of new programs, as well as advancements in existing ones, are presented. Finally, the option to validate the result by an independent procedure, a high quality ab initio quantum mech. calcn., is discussed.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2cXnsVKltr0%253D&md5=7d7486d14d46798ede0aa853faf013f9
19. 19
  Reymond, J. L.; Ruddigkeit, L.; Blum, L. C.; Van Deursen, R. The enumeration of chemical space Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 717– 733
  [Crossref], [CAS], Google Scholar
  19
  The enumeration of chemical space
  Reymond, Jean-Louis; Ruddigkeit, Lars; Blum, Lorenz; van Deursen, Ruud
  Wiley Interdisciplinary Reviews: Computational Molecular Science (2012), 2 (5), 717-733CODEN: WIRCAH; ISSN:1759-0884. (Wiley-Blackwell)
  A review. In the field of medicinal chem., the chem. space describes the ensemble of all org. mols. to be considered when searching for new drugs (estd. >1060 mols.), as well as the property spaces in which these mols. are placed for the sake of describing them. Mols. can be enumerated computationally by the millions, which was first undertaken in the field of computer-aided structure elucidation. Scoring the enumerated virtual libraries by virtual screening has recently become an attractive strategy to prioritize compds. for synthesis and testing. Enumeration methods include combinatorial linking of fragments, genetic algorithms based on cycles of enumeration and selection by ligand-based or target-based scoring functions, and exhaustive enumeration from first principles. The chem. space of mols. following simple rules of chem. stability and synthetic feasibility has been enumerated up to 13 atoms of C, N, O, Cl, S, forming the GDB-13 database with 977 million structures. The database has been organized in a 42-dimensional chem. space using mol. quantum nos. (MQN) as descriptors, which can be visualized by projection in two dimensions by principal component anal.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XhsFCmsr%252FM&md5=cda6dfd2d048e668455f52ee884b718a
20. 20
  Fink, T.; Bruggesser, H.; Reymond, J. L. Virtual exploration of the small-molecule chemical universe below 160 Da Angew. Chem., Int. Ed. Engl. 2005, 44, 1504– 1508
  [Crossref], [PubMed], Google Scholar
  There is no corresponding record for this reference.
21. 21
  Fink, T.; Reymond, J. L. Virtual exploration of the chemical universe up to 11 atoms of C, N, O, F: assembly of 26.4 million structures (110.9 million stereoisomers) and analysis for new ring systems, stereochemistry, physicochemical properties, compound classes, and drug discovery J. Chem. Inf. Model. 2007, 47, 342– 353
  [ACS Full Text ], [CAS], Google Scholar
  21
  Virtual Exploration of the Chemical Universe up to 11 Atoms of C, N, O, F: Assembly of 26.4 Million Structures (110.9 Million Stereoisomers) and Analysis for New Ring Systems, Stereochemistry, Physicochemical Properties, Compound Classes, and Drug Discovery
  Fink, Tobias; Reymond, Jean-Louis
  Journal of Chemical Information and Modeling (2007), 47 (2), 342-353CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  All mols. of up to 11 atoms of C, N, O, and F possible under consideration of simple valency, chem. stability, and synthetic feasibility rules were generated and collected in a database (GDB). GDB contains 26.4 million mols. (110.9 million stereoisomers), including three- and four-membered rings and triple bonds. By comparison, only 63 857 compds. of up to 11 atoms were found in public databases (a combination of PubChem, ChemACX, ChemSCX, NCI open database, and the Merck Index). A total of 538 of the 1208 ring systems in GDB are currently unknown in the CAS Registry and Beilstein databases in any carbon/heteroatom/multiple-bond combination or as a substructure. Over 70% of GDB mols. are chiral. Because of their small size, all compds. obey Lipinski's bioavailability rule. A total of 13.2 million compds. also follow Congreve's "Rule of 3" for lead-likeness. A Kohonen map trained with autocorrelation descriptors organizes GDB according to compd. classes and shows that leadlike compds. are most abundant in chiral regions of fused carbocycles and fused heterocycles. The projection of known compds. into this map indicates large uncharted areas of chem. space. The potential of GDB for drug discovery is illustrated by virtual screening for kinase inhibitors, G-protein coupled receptor ligands, and ion-channel modulators. The database is available from the author's Web page.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2sXhtVens7k%253D&md5=fe97c30ee8269de1e889648f4818f42f
22. 22
  Blum, L. C.; Reymond, J. L. 970 million druglike small molecules for virtual screening in the chemical universe database GDB-13 J. Am. Chem. Soc. 2009, 131, 8732– 8733
  [ACS Full Text ], [CAS], Google Scholar
  22
  970 Million Druglike Small Molecules for Virtual Screening in the Chemical Universe Database GDB-13
  Blum, Lorenz C.; Reymond, Jean-Louis
  Journal of the American Chemical Society (2009), 131 (25), 8732-8733CODEN: JACSAT; ISSN:0002-7863. (American Chemical Society)
  GDB-13 enumerates small org. mols. contg. up to 13 atoms of C, N, O, S, and Cl following simple chem. stability and synthetic feasibility rules. With 977 468 314 structures, GDB-13 is the largest publicly available small org. mol. database to date.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXmvFWru7k%253D&md5=22c200e887a6480b19a73852ca0a3435
23. 23
  Blum, L. C.; van Deursen, R.; Reymond, J. L. Visualisation and subsets of the chemical universe database GDB-13 for virtual screening J. Comput.-Aided Mol. Des. 2011, 25, 637– 647
  [Crossref], [PubMed], [CAS], Google Scholar
  23
  Visualisation and subsets of the chemical universe database GDB-13 for virtual screening
  Blum, Lorenz C.; Deursen, Ruud; Reymond, Jean-Louis
  Journal of Computer-Aided Molecular Design (2011), 25 (7), 637-647CODEN: JCADEQ; ISSN:0920-654X. (Springer)
  The chem. universe database GDB-13, which enumerates 977 million org. mols. up to 13 atoms of C, N, O, S and Cl following simple chem. stability and synthetic feasibility rules, represents a vast reservoir for new fragments. GDB-13 was classified using the MQN-system discussed in the preceding paper for the anal. of PubChem fragments. Two hundred and fifty-five subsets of GDB-13 were generated by the combinatorial use of eight restrictive criteria, including fragment-like ("rule of three") and scaffold-like (no acyclic carbon atoms) filters. Virtual screening for analogs of 15 com. drugs of 13 non-hydrogen atoms or less shows that retrieving MQN-neighbors of a query mol. from GDB-13 or its subsets provides on av. a 38-fold enrichment in structural analogs (Daylight-type substructure fingerprint Tanimoto TSF > 0.7), and a 75-fold enrichment in shape-similar analogs (ROCS TanimotoCombo score > 1.4). An MQN-searchable version of GDB-13 is provided at www.ghb.unibe.ch.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhtVSqtbbN&md5=45f426cafa64b3bb06fc11e68a1b4022
24. 24
  Nguyen, K. T.; Syed, S.; Urwyler, S.; Bertrand, S.; Bertrand, D.; Reymond, J. L. Discovery of NMDA glycine site inhibitors from the chemical universe database GDB ChemMedChem 2008, 3, 1520– 1524
  [Crossref], [PubMed], [CAS], Google Scholar
  24
  Discovery of NMDA glycine site inhibitors from the chemical universe database GDB
  Nguyen, Kong Thong; Syed, Salahuddin; Urwyler, Stephan; Bertrand, Sonia; Bertrand, Daniel; Reymond, Jean-Louis
  ChemMedChem (2008), 3 (10), 1520-1524CODEN: CHEMGX; ISSN:1860-7179. (Wiley-VCH Verlag GmbH & Co. KGaA)
  Using virtual screening tools, promising small mol. ligands of the NMDA receptor glycine site are identified for subsequent synthesis.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXhtlOgt73P&md5=d240511779046f0794b75e35f52c42cc
25. 25
  Nguyen, K. T.; Luethi, E.; Syed, S.; Urwyler, S.; Bertrand, S.; Bertrand, D.; Reymond, J. L. 3-(aminomethyl)piperazine-2,5-dione as a novel NMDA glycine site inhibitor from the chemical universe database GDB Bioorg. Med. Chem. Lett. 2009, 19, 3832– 3835
  [Crossref], [PubMed], Google Scholar
  There is no corresponding record for this reference.
26. 26
  Garcia-Delgado, N.; Bertrand, S.; Nguyen, K. T.; van Deursen, R.; Bertrand, D.; Reymond, J.-L. Exploring a7-nicotinic receptor ligand diversity by scaffold enumeration from the Chemical Universe Database GDB ACS Med. Chem. Lett. 2010, 1, 422– 426
  [ACS Full Text ], Google Scholar
  There is no corresponding record for this reference.
27. 27
  Luethi, E.; Nguyen, K. T.; Burzle, M.; Blum, L. C.; Suzuki, Y.; Hediger, M.; Reymond, J. L. Identification of selective norbornane-type aspartate analogue inhibitors of the glutamate transporter 1 (GLT-1) from the chemical universe generated database (GDB) J. Med. Chem. 2010, 53, 7236– 7250
  [ACS Full Text ], [CAS], Google Scholar
  27
  Identification of Selective Norbornane-Type Aspartate Analogue Inhibitors of the Glutamate Transporter 1 (GLT-1) from the Chemical Universe Generated Database (GDB)
  Luethi, Erika; Nguyen, Kong T.; Burzle, Marc; Blum, Lorenz C.; Suzuki, Yoshiro; Hediger, Matthias; Reymond, Jean-Louis
  Journal of Medicinal Chemistry (2010), 53 (19), 7236-7250CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  A variety of conformationally constrained aspartate and glutamate analogs inhibit the glutamate transporter 1 (GLT-1, also known as EAAT2). To expand the search for such analogs, a virtual library of aliph. aspartate and glutamate analogs was generated starting from the chem. universe database GDB-11, which contains 26.4 million possible mols. up to 11 atoms of C, N, O, F, resulting in 101026 aspartate analogs and 151285 glutamate analogs. Virtual screening was realized by high-throughput docking to the glutamate binding site of the glutamate transporter homolog from Pyrococcus horikoshii (PDB code: 1XFH) using Autodock. Norbornane-type aspartate analogs were selected from the top-scoring virtual hits and synthesized. Testing and optimization led to the identification of (1R*,2R*,3S*,4R*,6R*)-2-amino-6-phenethyl-bicyclo[2.2.1]heptane-2,3-dicarboxylic acid (I) as a new inhibitor of GLT-1 with IC50 = 1.4 μM against GLT-1 and no inhibition of the related transporter EAAC1. The systematic diversification of known ligands by enumeration with help of GDB followed by virtual screening, synthesis, and testing as exemplified here provides a general strategy for drug discovery.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhtFWlsLvF&md5=3834d28f487e11ead1a8a17698587935
28. 28
  Blum, L. C.; van Deursen, R.; Bertrand, S.; Mayer, M.; Burgi, J. J.; Bertrand, D.; Reymond, J. L. Discovery of alpha7-nicotinic receptor ligands by virtual screening of the Chemical Universe Database GDB-13 J. Chem. Inf. Model. 2011, 51, 3105– 3112
  [ACS Full Text ], [CAS], Google Scholar
  28
  Discovery of α7-Nicotinic Receptor Ligands by Virtual Screening of the Chemical Universe Database GDB-13
  Blum, Lorenz C.; van Deursen, Ruud; Bertrand, Sonia; Mayer, Milena; Burgi, Justus J.; Bertrand, Daniel; Reymond, Jean-Louis
  Journal of Chemical Information and Modeling (2011), 51 (12), 3105-3112CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  The chem. universe database GDB-13 enumerates 977 million org. mols. up to 13 atoms of C, N, O, Cl, and S that are virtually possible following simple rules for chem. stability and synthetic feasibility. Analogs of nicotine were identified in GDB-13 using the city-block distance in MQN-space (CBDMQN) as a similarity measure, combined with a restriction eliminating problematic structural elements. The search was carried out with a Web browser available at www.gdb.unibe.ch. This virtual screening procedure selected 31 504 analogs of nicotine from GDB-13, from which 48 were known nicotinic ligands reported in Chembl. An addnl. 60 virtual screening hits were purchased and tested for modulation of the acetylcholine signal at the human α7 nAChR expressed in Xenopus oocytes, which led to the identification of three previously unknown inhibitors. These expts. demonstrate for the first time the use of GDB-13 for ligand discovery.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhsVGru7vK&md5=c176859de6f18cc54b2f25eee42ab79a
29. 29
  Brethous, L.; Garcia-Delgado, N.; Schwartz, J.; Bertrand, S.; Bertrand, D.; Reymond, J. L. Synthesis and nicotinic receptor activity of chemical space analogues of N-(3R)-1-azabicyclo[2.2.2]oct-3-yl-4-chlorobenzamide (PNU-282,987) and 1,4-diazabicyclo[3.2.2]nonane-4-carboxylic acid 4-bromophenyl ester (SSR180711) J. Med. Chem. 2012, 55, 4605– 4618
  [ACS Full Text ], Google Scholar
  There is no corresponding record for this reference.
30. 30
  Reymond, J. L.; Awale, M. Exploring chemical space for drug discovery using the Chemical Universe Database ACS Chem. Neurosci. 2012, 3, 649– 657
  [ACS Full Text ], [CAS], Google Scholar
  30
  Exploring Chemical Space for Drug Discovery Using the Chemical Universe Database
  Reymond, Jean-Louis; Awale, Mahendra
  ACS Chemical Neuroscience (2012), 3 (9), 649-657CODEN: ACNCDM; ISSN:1948-7193. (American Chemical Society)
  Herein we review our recent efforts in searching for bioactive ligands by enumeration and virtual screening of the unknown chem. space of small mols. Enumeration from first principles shows that almost all small mols. (>99.9%) have never been synthesized and are still available to be prepd. and tested. We discuss open access sources of mols., the classification and representation of chem. space using mol. quantum nos. (MQN), its exhaustive enumeration in form of the chem. universe generated databases (GDB), and examples of using these databases for prospective drug discovery. MQN-searchable GDB, PubChem, and DrugBank are freely accessible at www.gdb.unibe.ch.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC38XmtValu70%253D&md5=bbbe3c0931328f3796ce999189374864
31. 31
  Foloppe, N. The benefits of constructing leads from fragment hits Future Med. Chem. 2011, 3, 1111– 1115
  [Crossref], [PubMed], [CAS], Google Scholar
  31
  The benefits of constructing leads from fragment hits
  Foloppe, N.
  Future Medicinal Chemistry (2011), 3 (9), 1111-1115CODEN: FMCUA7; ISSN:1756-8919. (Future Science Ltd.)
  A review. Fragments'' refer to particularly small mol. starting points in medicinal chem. The small size of fragments requires adapted techniques for their screening and subsequent elaboration. The detection of the weak binding affinity of fragments for their target, and assocd. screening issues, have been debated at length. Since it is now clear that fragments can be developed into clin. candidates, the discussion is shifting to the design of good-quality lead compds. from fragment hits. The increasing ability to control and tailor this construction process highlights the potential benefits of fragment-based drug discovery.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXpsFeisLY%253D&md5=915652874047d19b21987cccf52c0df7
32. 32
  Teague, S. J.; Davis, A. M.; Leeson, P. D.; Oprea, T. The design of leadlike combinatorial libraries Angew. Chem., Int. Ed. Engl. 1999, 38, 3743– 3748
  [Crossref], Google Scholar
  There is no corresponding record for this reference.
33. 33
  Wang, Y.; Xiao, J.; Suzek, T. O.; Zhang, J.; Wang, J.; Bryant, S. H. PubChem: a public information system for analyzing bioactivities of small molecules Nucleic Acids Res. 2009, 37, W623– W633
  [Crossref], [PubMed], [CAS], Google Scholar
  33
  PubChem: a public information system for analyzing bioactivities of small molecules
  Wang, Yanli; Xiao, Jewen; Suzek, Tugba O.; Zhang, Jian; Wang, Jiyao; Bryant, Stephen H.
  Nucleic Acids Research (2009), 37 (Web Server), W623-W633CODEN: NARHAD; ISSN:0305-1048. (Oxford University Press)
  PubChem (http://pubchem.ncbi.nlm.nih.gov) is a public repository for biol. properties of small mols. hosted by the US National Institutes of Health (NIH). PubChem BioAssay database currently contains biol. test results for more than 700 000 compds. The goal of PubChem is to make this information easily accessible to biomedical researchers. In this work, we present a set of web servers to facilitate and optimize the utility of biol. activity information within PubChem. These web-based services provide tools for rapid data retrieval, integration and comparison of biol. screening results, exploratory structure-activity anal., and target selectivity examn. This article reviews these bioactivity anal. tools and discusses their uses. Most of the tools described in this work can be directly accessed at http://pubchem.ncbi.nlm.nih.gov/assay/. URLs for accessing other tools described in this work are specified individually.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXosFSktL8%253D&md5=11d50a1ff4d9b353728e1e03ee2e33ca
34. 34
  Gaulton, A.; Bellis, L. J.; Bento, A. P.; Chambers, J.; Davies, M.; Hersey, A.; Light, Y.; McGlinchey, S.; Michalovich, D.; Al-Lazikani, B.; Overington, J. P. ChEMBL: a large-scale bioactivity database for drug discovery Nucleic Acids Res. 2012, 40, D1100– D1107
  [Crossref], [PubMed], [CAS], Google Scholar
  34
  ChEMBL: a large-scale bioactivity database for drug discovery
  Gaulton, Anna; Bellis, Louisa J.; Bento, A. Patricia; Chambers, Jon; Davies, Mark; Hersey, Anne; Light, Yvonne; McGlinchey, Shaun; Michalovich, David; Al-Lazikani, Bissan; Overington, John P.
  Nucleic Acids Research (2012), 40 (D1), D1100-D1107CODEN: NARHAD; ISSN:0305-1048. (Oxford University Press)
  ChEMBL is an Open Data database contg. binding, functional and ADMET information for a large no. of drug-like bioactive compds. These data are manually abstracted from the primary published literature on a regular basis, then further curated and standardized to maximize their quality and utility across a wide range of chem. biol. and drug-discovery research problems. Currently, the database contains 5.4 million bioactivity measurements for more than 1 million compds. and 5200 protein targets. Access is available through a web-based interface, data downloads and web services at: https://www.ebi.ac.uk/chembldb.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhs12htbjN&md5=aedf7793e1ca54b6a4fa272ea3ef7d0e
35. 35
  Knox, C.; Law, V.; Jewison, T.; Liu, P.; Ly, S.; Frolkis, A.; Pon, A.; Banco, K.; Mak, C.; Neveu, V.; Djoumbou, Y.; Eisner, R.; Guo, A. C.; Wishart, D. S. DrugBank 3.0: a comprehensive resource for ‘Omics’ research on drugs Nucleic Acids Res. 2011, 39, D1035– D1041
  [Crossref], Google Scholar
  There is no corresponding record for this reference.
36. 36
  McKay, B. D. Practical graph isomorphism Congressus Numerantium 1981, 30, 45– 87
  Google Scholar
  There is no corresponding record for this reference.
37. 37
  Rishton, G. M. Reactive compounds and in vitro false positives in HTS Drug Discovery Today 1997, 2, 382– 384
  [Crossref], [CAS], Google Scholar
  37
  Reactive compounds and in vitro false positives in HTS
  Rishton, Gilbert M.
  Drug Discovery Today (1997), 2 (9), 382-384CODEN: DDTOFS; ISSN:1359-6446. (Elsevier)
  A review without refs. An important component of the successful high-throughput screening (HTS) strategy in drug discovery is the ability to assess HTS structure-activity data, and to distinguish between promising drug leads and the many useless false positives that can plague screening efforts. The author discusses simple chem. guidelines for the evaluation of "positives" in biochem. screens, with the aim of selecting stable, non-covalent binders (ligands) and eliminating protein-reactive compds. (reagents) from consideration as drug leads at an early stage.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK2sXlvFeksL4%253D&md5=b43a00783098a324ab3dc4d5f3704013
38. 38
  Rishton, G. M. Nonleadlikeness and leadlikeness in biochemical screening Drug Discovery Today 2003, 8, 86– 96
  [Crossref], [PubMed], [CAS], Google Scholar
  38
  Nonleadlikeness and leadlikeness in biochemical screening
  Rishton Gilbert M
  Drug discovery today (2003), 8 (2), 86-96 ISSN:1359-6446.
  Biochemical assays have largely supplanted functional biological assays as drug screening tools in the early stages of drug discovery. The de-selection of compounds that are 'nonleadlike' binders (and bonders) and the proactive selection of those compounds that are 'leadlike' in their binding to the target are vital components of the screening effort. The physiochemical properties of leadlikeness and the surprising differences between those properties and the now classical definitions of druglikeness are becoming apparent.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A280%3ADC%252BD3s%252Fls1SjtQ%253D%253D&md5=5e34cece6f751fa727f949c3526aa95c
39. 39
  Rush, T. S., III; Grant, J. A.; Mosyak, L.; Nicholls, A. A shape-based 3-D scaffold hopping method and its application to a bacterial protein-protein interaction J. Med. Chem. 2005, 48, 1489– 1495
  [ACS Full Text ], [CAS], Google Scholar
  39
  A Shape-Based 3-D Scaffold Hopping Method and Its Application to a Bacterial Protein-Protein Interaction
  Rush, Thomas S., III; Grant, J. Andrew; Mosyak, Lidia; Nicholls, Anthony
  Journal of Medicinal Chemistry (2005), 48 (5), 1489-1495CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  In this paper, the authors describe the first prospective application of the shape-comparison program ROCS (Rapid Overlay of Chem. Structures) to find new scaffolds for small mol. inhibitors of the ZipA-FtsZ protein-protein interaction, a proposed antibacterial target. The shape comparisons are made relative to the crystallog. detd., bioactive conformation of a high-throughput screening (HTS) hit. The use of ROCS led to the identification of a set of novel, weakly binding inhibitors with scaffolds presenting synthetic opportunities to further optimize biol. affinity and lacking development issues assocd. with the HTS lead. These ROCS-identified scaffolds would have been missed using other structural similarity approaches such as ISIS 2D fingerprints. X-ray crystallog. anal. of one of the new inhibitors bound to ZipA reveals that the shape comparison approach very accurately predicted the binding mode. These exptl. results validate this use of ROCS for chemotype switching or "lead hopping" and suggest that it is of general interest for lead identification in drug discovery endeavors.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD2MXht1Ols78%253D&md5=05b6e54a657a3b8c768e63852d871ef6
40. 40
  Nicholls, A.; McGaughey, G. B.; Sheridan, R. P.; Good, A. C.; Warren, G.; Mathieu, M.; Muchmore, S. W.; Brown, S. P.; Grant, J. A.; Haigh, J. A.; Nevins, N.; Jain, A. N.; Kelley, B. Molecular shape and medicinal chemistry: a perspective J. Med. Chem. 2010, 53, 3862– 3886
  [ACS Full Text ], [CAS], Google Scholar
  40
  Molecular Shape and Medicinal Chemistry: A Perspective
  Nicholls, Anthony; McGaughey, Georgia B.; Sheridan, Robert P.; Good, Andrew C.; Warren, Gregory; Mathieu, Magali; Muchmore, Steven W.; Brown, Scott P.; Grant, J. Andrew; Haigh, James A.; Nevins, Neysa; Jain, Ajay N.; Kelley, Brian
  Journal of Medicinal Chemistry (2010), 53 (10), 3862-3886CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  A review article with 111 refs. summarized perspectives of mol. shape and medicinal chem. in drug screening.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhvF2kt7k%253D&md5=85664344e13872527a3dfb2296d34864
41. 41
  Sauer, W. H.; Schwarz, M. K. Molecular shape diversity of combinatorial libraries: a prerequisite for broad bioactivity J. Chem. Inf. Comput. Sci. 2003, 43, 987– 1003
  [ACS Full Text ], [CAS], Google Scholar
  41
  Molecular Shape Diversity of Combinatorial Libraries: A Prerequisite for Broad Bioactivity
  Sauer, Wolfgang H. B.; Schwarz, Matthias K.
  Journal of Chemical Information and Computer Sciences (2003), 43 (3), 987-1003CODEN: JCISD8; ISSN:0095-2338. (American Chemical Society)
  A computational method to rapidly assess and visualize the diversity in mol. shape assocd. with a given compd. set has been developed. Normalized ratios of principal moments of inertia are plotted into two-dimensional triangular graphs and then used to compare the shape space covered by different compd. sets, such as combinatorial libraries of varying size and compn. We have further developed a computational method to analyze interset similarity in terms of shape space coverage, which allows the shape redundancy between the different subsets of a given compd. collection to be analyzed in a quant. way. The shape space coverage has been found to originate mainly from the nature and the 3D-geometry (but not the size) of the central scaffold, while the no. and nature of the peripheral substituents and conformational aspects were shown to be of minor importance. Substantial shape space coverage has been correlated with broad biol. activity by applying the same shape anal. to collections of known bioactive compds., such as MDDR and the GOLD-set. The aggregate of our results corroborates the intuitive notion that mol. shape is intimately linked to biol. activity and that a high degree of shape (hence scaffold) diversity in screening collections will increase the odds of addressing a broad range of biol. targets.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXhvF2muro%253D&md5=c07401742457e76b066bed4b43feca1d
42. 42
  Lovering, F.; Bikker, J.; Humblet, C. Escape from flatland: increasing saturation as an approach to improving clinical success J. Med. Chem. 2009, 52, 6752– 6756
  [ACS Full Text ], [CAS], Google Scholar
  42
  Escape from Flatland: Increasing Saturation as an Approach to Improving Clinical Success
  Lovering, Frank; Bikker, Jack; Humblet, Christine
  Journal of Medicinal Chemistry (2009), 52 (21), 6752-6756CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  The medicinal chem. community has become increasingly aware of the value of tracking calcd. phys. properties such as mol. wt., topol. polar surface area, rotatable bonds, and hydrogen bond donors and acceptors. The authors hypothesized that the shift to high-throughput synthetic practices over the past decade may be another factor that may predispose mols. to fail by steering discovery efforts toward achiral, arom. compds. The authors have proposed two simple and interpretable measures of the complexity of mols. prepd. as potential drug candidates. The first is carbon bond satn. as defined by fraction Sp3 (Fsp3) where Fsp3 = (no. of Sp3 hybridized carbons/total carbon count). The second is simply whether a chiral carbon exists in the mol. The authors demonstrate that both complexity (as measured by Fsp3) and the presence of chiral centers correlate with success as compds. transition from discovery, through clin. testing, to drugs. To explain these observations, the authors further demonstrate that satn. correlates with soly., an exptl. phys. property important to success in the drug discovery setting.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1MXht1KjtLvN&md5=4ca92c30c17c53d77ad376719bad951e
43. 43
  Ritchie, T. J.; Macdonald, S. J.; Young, R. J.; Pickett, S. D. The impact of aromatic ring count on compound developability: further insights by examining carbo- and hetero-aromatic and -aliphatic ring types Drug Discovery Today 2011, 16, 164– 171
  [Crossref], [PubMed], [CAS], Google Scholar
  43
  The impact of aromatic ring count on compound developability: further insights by examining carbo- and hetero-aromatic and -aliphatic ring types
  Ritchie, Timothy J.; MacDonald, Simon J. F.; Young, Robert J.; Pickett, Stephen D.
  Drug Discovery Today (2011), 16 (3/4), 164-171CODEN: DDTOFS; ISSN:1359-6446. (Elsevier B.V.)
  A review. The impact of carboarom., heteroarom., carboaliph. and heteroaliph. ring counts and fused arom. ring count on several developability measures (soly., lipophilicity, protein binding, P 450 inhibition and hERG binding) is the topic for this review article. Recent results indicate that increasing ring counts have detrimental effects on developability in the order carboaroms. » heteroaroms. > carboaliphatics > heteroaliphatics, with heteroaliphatics exerting a beneficial effect in many cases. Increasing arom. ring count exerts effects on several developability parameters that are lipophilicity- and size-independent, and fused arom. systems have a beneficial effect relative to their nonfused counterparts. Increasing arom. ring count has a detrimental effect on human bioavailability parameters, and heteroarom. ring count (but not other ring counts) has increased over time in marketed oral drugs.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXhvVyqsr0%253D&md5=8e2d0e4e499d2f07d66ca2f385defb2d
44. 44
  Clemons, P. A.; Bodycombe, N. E.; Carrinski, H. A.; Wilson, J. A.; Shamji, A. F.; Wagner, B. K.; Koehler, A. N.; Schreiber, S. L. Small molecules of different origins have distinct distributions of structural complexity that correlate with protein-binding profiles Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 18787– 18792
  [Crossref], [PubMed], [CAS], Google Scholar
  44
  Small molecules of different origins have distinct distributions of structural complexity that correlate with protein-binding profiles
  Clemons, Paul A.; Bodycombe, Nicole E.; Carrinski, Hyman A.; Wilson, J. Anthony; Shamji, Alykhan F.; Wagner, Bridget K.; Koehler, Angela N.; Schreiber, Stuart L.
  Proceedings of the National Academy of Sciences of the United States of America (2010), 107 (44), 18787-18792, S18787/1-S18787/5CODEN: PNASA6; ISSN:0027-8424. (National Academy of Sciences)
  Using a diverse collection of small mols. generated from a variety of sources, we measured protein-binding activities of each individual compd. against each of 100 diverse (sequence-unrelated) proteins using small-mol. microarrays. We also analyzed structural features, including complexity, of the small mols. We found that compds. from different sources (com., academic, natural) have different protein-binding behaviors and that these behaviors correlate with general trends in stereochem. and shape descriptors for these compd. collections. Increasing the content of sp3-hybridized and stereogenic atoms relative to compds. from com. sources, which comprise the majority of current screening collections, improved binding selectivity and frequency. The results suggest structural features that synthetic chemists can target when synthesizing screening collections for biol. discovery. Because binding proteins selectively can be a key feature of high-value probes and drugs, synthesizing com- pounds having features identified in this study may result in improved performance of screening collections.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3cXhsVCqsb%252FN&md5=198e298ca5391d7dfd91184680e3c2ec
45. 45
  Clemons, P. A.; Wilson, J. A.; Dancik, V.; Muller, S.; Carrinski, H. A.; Wagner, B. K.; Koehler, A. N.; Schreiber, S. L. Quantifying structure and performance diversity for sets of small molecules comprising small-molecule screening collections Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 6817– 6822
  [Crossref], [PubMed], [CAS], Google Scholar
  45
  Quantifying structure and performance diversity for sets of small molecules comprising small-molecule screening collections
  Clemons, Paul A.; Wilson, J. Anthony; Dancik, Vlado; Muller, Sandrine; Carrinski, Hyman A.; Wagner, Bridget K.; Koehler, Angela N.; Schreiber, Stuart L.
  Proceedings of the National Academy of Sciences of the United States of America (2011), 108 (17), 6817-6822, S6817/1-S6817/8CODEN: PNASA6; ISSN:0027-8424. (National Academy of Sciences)
  Using a diverse collection of small mols. we recently found that compd. sets from different sources (com.; academic; natural) have different protein-binding behaviors, and these behaviors correlate with trends in stereochem. complexity for these compd. sets. These results lend insight into structural features that synthetic chemists might target when synthesizing screening collections for biol. discovery. We report extensive characterization of structural properties and diversity of biol. performance for these compds. and expand comparative analyses to include physicochem. properties and three-dimensional shapes of predicted conformers. The results highlight addnl. similarities and differences between the sets, but also the dependence of such comparisons on the choice of mol. descriptors. Using a protein-binding dataset, we introduce an information-theoretic measure to assess diversity of performance with a constraint on specificity. Rather than relying on finding individual active compds., this measure allows rational judgment of compd. subsets as groups. We also apply this measure to publicly available data from ChemBank for the same compd. sets across a diverse group of functional assays. We find that performance diversity of compd. sets is relatively stable across a range of property values as judged by this measure, both in protein-binding studies and functional assays. Because building screening collections with improved performance depends on efficient use of synthetic org. chem. resources, these studies illustrate an important quant. framework to help prioritize choices made in building such collections.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC3MXlslyru7c%253D&md5=d83343549289415060ea55d45118d31f
46. 46
  Sadowski, J.; Gasteiger, J. From atoms and bonds to 3-dimensional atomic coordinates - automatic model builders Chem. Rev. 1993, 93, 2567– 2581
  [ACS Full Text ], [CAS], Google Scholar
  46
  From atoms and bonds to three-dimensional atomic coordinates: automatic model builders
  Sadowski, Jens; Gasteiger, Johann
  Chemical Reviews (Washington, DC, United States) (1993), 93 (7), 2567-81CODEN: CHREAY; ISSN:0009-2665.
  A review with ∼75 refs. in which WIZARD, COBRA, CONCORD, and CORINA are discussed.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK3sXmt1GgsLs%253D&md5=fcdb50dfbd981c8f122da06e06e62e27
47. 47
  Bemis, G. W.; Murcko, M. A. The properties of known drugs. 1. Molecular frameworks J. Med. Chem. 1996, 39, 2887– 2893
  [ACS Full Text ], [CAS], Google Scholar
  47
  The Properties of Known Drugs. 1. Molecular Frameworks
  Bemis, Guy W.; Murcko, Mark A.
  Journal of Medicinal Chemistry (1996), 39 (15), 2887-2893CODEN: JMCMAR; ISSN:0022-2623. (American Chemical Society)
  To better understand the common features present in drug mols., we use shape description methods to analyze a database of com. available drugs and prep. a list of common drug shapes. A useful way of organizing this structural data is to group the atoms of each drug mol. into ring, linker, framework, and side chain atoms. On the basis of the two-dimensional mol. structures (without regard to atom type, hybridization, and bond order), there are 1179 different frameworks among the 5120 compds. analyzed. However, the shapes of half of the drugs in the database are described by the 32 most frequently occurring frameworks. This suggests that the diversity of shapes in the set of known drugs is extremely low. In our second method of anal., in which atom type, hybridization, and bond order are considered, more diversity is seen; there are 2506 different frameworks among the 5120 compds. in the database, and the most frequently occurring 42 frameworks account for only one-fourth of the drugs. We discuss the possible interpretations of these findings and the way they may be used to guide future drug discovery research.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK28XjvVejtro%253D&md5=5e2c4fdfea9434456a0cca83de4185b3

PDF [ 3MB]

All Types

Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17

Article Views

Altmetric

Citations

Abstract

Introduction

Results and Discussion

Enumeration

Figure 1

Comparing GDB-17 with PubChem, ChEMBL, and DrugBank

Figure 2

Figure 3

Figure 4

Small Rings, Topology, and Compound Categories

Polarity and Leadlikeness

Figure 5

Molecular Shape

Figure 6

Figure 7

Stereochemistry

Figure 8

Novelty and Scaffolds

Figure 9

Conclusion

Methods

General

Enumeration

Graphs

Hydrocarbons

Skeletons

CNO Molecules

Postprocessing for Oximes, Nitro, CF3, Halogens, and Sulfur

Shape Analysis

Stereoisomer Counting

Distribution

Author Information

Acknowledgment

References

Cited By

Abstract

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

References

STEP 1:

STEP 2:

Postprocessing for Oximes, Nitro, CF₃, Halogens, and Sulfur