Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

You’ve supercharged your research process with ACS and Mendeley!

Click to create an ACS ID

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Your Mendeley pairing has expired. Please reconnect
ACS Publications. Most Trusted. Most Cited. Most Read
My Activity

Figure 1Loading Img

Comparison of Cellular Morphological Descriptors and Molecular Fingerprints for the Prediction of Cytotoxicity- and Proliferation-Related Assays

Cite this: Chem. Res. Toxicol. 2021, 34, 2, 422–437
Publication Date (Web):February 1, 2021
Copyright © 2021 American Chemical Society

    Article Views





    Other access options
    Supporting Info (2)»


    Abstract Image

    Cell morphology features, such as those from the Cell Painting assay, can be generated at relatively low costs and represent versatile biological descriptors of a system and thereby compound response. In this study, we explored cell morphology descriptors and molecular fingerprints, separately and in combination, for the prediction of cytotoxicity- and proliferation-related in vitro assay endpoints. We selected 135 compounds from the MoleculeNet ToxCast benchmark data set which were annotated with Cell Painting readouts, where the relatively small size of the data set is due to the overlap of required annotations. We trained Random Forest classification models using nested cross-validation and Cell Painting descriptors, Morgan and ErG fingerprints, and their combinations. While using leave-one-cluster-out cross-validation (with clusters based on physicochemical descriptors), models using Cell Painting descriptors achieved higher average performance over all assays (Balanced Accuracy of 0.65, Matthews Correlation Coefficient of 0.28, and AUC-ROC of 0.71) compared to models using ErG fingerprints (BA 0.55, MCC 0.09, and AUC-ROC 0.60) and Morgan fingerprints alone (BA 0.54, MCC 0.06, and AUC-ROC 0.56). While using random shuffle splits, the combination of Cell Painting descriptors with ErG and Morgan fingerprints further improved balanced accuracy on average by 8.9% (in 9 out of 12 assays) and 23.4% (in 8 out of 12 assays) compared to using only ErG and Morgan fingerprints, respectively. Regarding feature importance, Cell Painting descriptors related to nuclei texture, granularity of cells, and cytoplasm as well as cell neighbors and radial distributions were identified to be most contributing, which is plausible given the endpoint considered. We conclude that cell morphological descriptors contain complementary information to molecular fingerprints which can be used to improve the performance of predictive cytotoxicity models, in particular in areas of novel structural space.

    Read this article

    To access this article, please review the available access options below.

    Get instant access

    Purchase Access

    Read this article for 48 hours. Check out below using your ACS ID or as a guest.


    Access through Your Institution

    You may have access to this article through your institution.

    Your institution does not have access to this content. You can change your affiliated institution below.

    Supporting Information

    Jump To

    The Supporting Information is available free of charge at

    • Supplementary Table S1, description of data set; Figure S1, distribution of active and inactive compounds among selected endpoints; Figure S2, distribution of chemical space of 135 compounds used in this study compared to all compounds present in same 12 assays from ToxCast; Figure S3, pairwise Tanimoto similarity (Tc) of compounds in data set used in this study using (a) 166 public MACCS keys and (b) Morgan fingerprints of radius 2 and 2048 nbits and contour graph of Tanimoto similarity coefficients to show molecule similarity using (c) 166 public MACCS keys and (d) Morgan fingerprints of radius 2 and 2048 nbits; Table S2, hyperparameter search spaces for Random Forest model; Table S3, explained variance using first two principal components PC1 and PC2 from PCA and silhouette scores for 5 clusters among selected endpoints; Table S4, aggregated metrics for Random Forest model using shuffle stratified splitting; Table S5, cluster-averaged metrics for Random Forest model using leave-one-cluster-out splitting; Figure S4, Area Under Curve-Receiver Operating Characteristic (AUC-ROC) performance of cluster averaged models for 12 assays; Figure S5, Mathew’s correlation constant (MCC) performance of cluster averaged models for 12 assays; Figure S6, Balanced Accuracy (BA) performance of cluster averaged models for 12 assays; and Table S7, feature importance of Cell Painting descriptors (PDF)

    • Table S6, confusion matrices of Random Forest Models (XLSX)

    Terms & Conditions

    Electronic Supporting Information files are available without a subscription to ACS Web Editions. The American Chemical Society holds a copyright ownership interest in any copyrightable Supporting Information. Files available from the ACS website may be downloaded for personal use only. Users are not otherwise permitted to reproduce, republish, redistribute, or sell any Supporting Information from the ACS website, either in whole or in part, in either machine-readable form or any other form without permission from the American Chemical Society. For permission to reproduce, republish and redistribute this material, requesters must process their own requests via the RightsLink permission system. Information about how to use the RightsLink permission system can be found at

    Cited By

    This article is cited by 22 publications.

    1. Srijit Seal, Ola Spjuth, Layla Hosseini-Gerami, Miguel García-Ortegón, Shantanu Singh, Andreas Bender, Anne E. Carpenter. Insights into Drug Cardiotoxicity from Biological and Chemical Data: The First Public Classifiers for FDA Drug-Induced Cardiotoxicity Rank. Journal of Chemical Information and Modeling 2024, 64 (4) , 1172-1186.
    2. Robert I. Horne, Jared Wilson-Godber, Alicia González Díaz, Z. Faidon Brotzakis, Srijit Seal, Rebecca C. Gregory, Andrea Possenti, Sean Chia, Michele Vendruscolo. Using Generative Modeling to Endow with Potency Initially Inert Compounds with Good Bioavailability and Low Toxicity. Journal of Chemical Information and Modeling 2024, 64 (3) , 590-596.
    3. Vanille Lejal, Natacha Cerisier, David Rouquié, Olivier Taboureau. Assessment of Drug-Induced Liver Injury through Cell Morphology and Gene Expression Analysis. Chemical Research in Toxicology 2023, 36 (9) , 1456-1470.
    4. Marina Garcia de Lomana, Paula Andrea Marin Zapata, Floriane Montanari. Predicting the Mitochondrial Toxicity of Small Molecules: Insights from Mechanistic Assays and Cell Painting Data. Chemical Research in Toxicology 2023, 36 (7) , 1107-1120.
    5. Maria-Anna Trapotsi, Elizabeth Mouchet, Guy Williams, Tiziana Monteverde, Karolina Juhani, Riku Turkki, Filip Miljković, Anton Martinsson, Lewis Mervin, Kenneth R. Pryde, Erik Müllers, Ian Barrett, Ola Engkvist, Andreas Bender, Kevin Moreau. Cell Morphological Profiling Enables High-Throughput Screening for PROteolysis TArgeting Chimera (PROTAC) Phenotypic Signature. ACS Chemical Biology 2022, 17 (7) , 1733-1744.
    6. Nicole C. Kleinstreuer, (National Institute of Environmental Health Sciences)Igor V. Tetko, (Associate Editor)Weida Tong (Food and Drug Administration). Introduction to Special Issue: Computational Toxicology. Chemical Research in Toxicology 2021, 34 (2) , 171-175.
    7. Anjana S Desai, Anindita Bandopadhyaya, Aparna Ashok, Maneesha, Neeru Bhagat. Decoding characteristics of key physical properties in silver nanoparticles by attaining centroids for cytotoxicity prediction through data cleansing. Machine Learning: Science and Technology 2024, 5 (2) , 025059.
    8. Srijit Seal, Maria-Anna Trapotsi, Ola Spjuth, Shantanu Singh, Jordi Carreras-Puigvert, Nigel Greene, Andreas Bender, Anne E. Carpenter. A Decade in a Systematic Review: The Evolution and Impact of Cell Painting. 2024
    9. Shengkun Ni, Xiangtai Kong, Yingying Zhang, Zhengyang Chen, Zhaokun Wang, Zunyun Fu, Ruifeng Huo, Xiaochu Tong, Ning Qu, Xiaolong Wu, Kun Wang, Wei Zhang, Runze Zhang, Zimei Zhang, Jiangshan Shi, Yitian Wang, Ruirui Yang, Xutong Li, Sulin Zhang, Mingyue Zheng. Identify compound-protein interaction with knowledge graph embedding of perturbation transcriptomics. 2024
    10. Srijit Seal, Jordi Carreras-Puigvert, Shantanu Singh, Anne E. Carpenter, Ola Spjuth, Andreas Bender, . From pixels to phenotypes: Integrating image-based profiling with cell health data as BioMorph features improves interpretability. Molecular Biology of the Cell 2024, 35 (3)
    11. Guangyan Tian, Philip J Harrison, Akshai P Sreenivasan, Jordi Carreras-Puigvert, Ola Spjuth. Combining molecular and cell painting image data for mechanism of action prediction. Artificial Intelligence in the Life Sciences 2023, 3 , 100060.
    12. Srijit Seal, Hongbin Yang, Maria-Anna Trapotsi, Satvik Singh, Jordi Carreras-Puigvert, Ola Spjuth, Andreas Bender. Merging bioactivity predictions from cell morphology and chemical fingerprint models using similarity to training data. Journal of Cheminformatics 2023, 15 (1)
    13. Anika Liu, Srijit Seal, Hongbin Yang, Andreas Bender. Using chemical and biological data to predict drug toxicity. SLAS Discovery 2023, 28 (3) , 53-64.
    14. Natacha Cerisier, Bryan Dafniet, Anne Badel, Olivier Taboureau. Linking chemicals, genes and morphological perturbations to diseases. Toxicology and Applied Pharmacology 2023, 461 , 116407.
    15. Lijo John, Hridoy Jyoti Mahanta, Y. Soujanya, G. Narahari Sastry. Assessing machine learning approaches for predicting failures of investigational drug candidates during clinical trials. Computers in Biology and Medicine 2023, 153 , 106494.
    16. Mainak Chatterjee, Kunal Roy. Quantitative structure-activity relationships (QSARs) in medicinal chemistry. 2023, 3-38.
    17. Andi Alijagic, Nikolai Scherbak, Oleksandr Kotlyar, Patrik Karlsson, Xuying Wang, Inger Odnevall, Oldřich Benada, Ali Amiryousefi, Lena Andersson, Alexander Persson, Jenny Felth, Henrik Andersson, Maria Larsson, Alexander Hedbrant, Samira Salihovic, Tuulia Hyötyläinen, Dirk Repsilber, Eva Särndahl, Magnus Engwall. A Novel Nanosafety Approach Using Cell Painting, Metabolomics, and Lipidomics Captures the Cellular and Molecular Phenotypes Induced by the Unintentionally Formed Metal-Based (Nano)Particles. Cells 2023, 12 (2) , 281.
    18. Hongbin Yang, Olga Obrezanova, Amy Pointon, Will Stebbeds, Jo Francis, Kylie A. Beattie, Peter Clements, James S. Harvey, Graham F. Smith, Andreas Bender. Prediction of inotropic effect based on calcium transients in human iPSC-derived cardiomyocytes and machine learning. Toxicology and Applied Pharmacology 2023, 459 , 116342.
    19. Srijit Seal, Jordi Carreras-Puigvert, Maria-Anna Trapotsi, Hongbin Yang, Ola Spjuth, Andreas Bender. Integrating cell morphology with gene expression and chemical structure to aid mitochondrial toxicity detection. Communications Biology 2022, 5 (1)
    20. Jaeseong Jeong, Donghyeon Kim, Jinhee Choi. Application of ToxCast/Tox21 data for toxicity mechanism-based evaluation and prioritization of environmental chemicals: Perspective and limitations. Toxicology in Vitro 2022, 84 , 105451.
    21. Jonne Rietdijk, Tanya Aggarwal, Polina Georgieva, Maris Lapins, Jordi Carreras-Puigvert, Ola Spjuth. Morphological profiling of environmental chemicals enables efficient and untargeted exploration of combination effects. Science of The Total Environment 2022, 832 , 155058.
    22. Maria-Anna Trapotsi, Layla Hosseini-Gerami, Andreas Bender. Computational analyses of mechanism of action (MoA): data, methods and integration. RSC Chemical Biology 2022, 3 (2) , 170-200.
    23. Morgan Thomas, Andrew Boardman, Miguel Garcia-Ortegon, Hongbin Yang, Chris de Graaf, Andreas Bender. Applications of Artificial Intelligence in Drug Design: Opportunities and Challenges. 2022, 1-59.
    24. Jonne Rietdijk, Marianna Tampere, Aleksandra Pettke, Polina Georgiev, Maris Lapins, Ulrika Warpman-Berglund, Ola Spjuth, Marjo-Riitta Puumalainen, Jordi Carreras-Puigvert. A phenomics approach for antiviral drug discovery. BMC Biology 2021, 19 (1)