Predicting Compound Activity from Phenotypic Profiles and Chemical Structures

Overview

Journal Nat Commun

Specialty Biology

Date 2023 Apr 8

PMID 37031208

Authors

Nikita Moshkov

Tim Becker

Kevin Yang

Peter Horvath

Vlado Dancik

Bridget K Wagner

Paul A Clemons

Shantanu Singh

Anne E Carpenter

Juan C Caicedo

Affiliations

Soon will be listed here.

Abstract

Predicting assay results for compounds virtually using chemical structures and phenotypic profiles has the potential to reduce the time and resources of screens for drug discovery. Here, we evaluate the relative strength of three high-throughput data sources-chemical structures, imaging (Cell Painting), and gene-expression profiles (L1000)-to predict compound bioactivity using a historical collection of 16,170 compounds tested in 270 assays for a total of 585,439 readouts. All three data modalities can predict compound activity for 6-10% of assays, and in combination they predict 21% of assays with high accuracy, which is a 2 to 3 times higher success rate than using a single modality alone. In practice, the accuracy of predictors could be lower and still be useful, increasing the assays that can be predicted from 37% with chemical structures alone up to 64% when combined with phenotypic data. Our study shows that unbiased phenotypic profiling can be leveraged to enhance compound bioactivity prediction to accelerate the early stages of the drug-discovery process.

Citing Articles

CPHNet: a novel pipeline for anti-HAPE drug screening via deep learning-based Cell Painting scoring.

Sun D, Yang X, Huang C, Bai Z, Shen P, Ni Z Respir Res. 2025; 26(1):91.

PMID: 40057746 PMC: 11890554. DOI: 10.1186/s12931-025-03173-1.

Evaluating feature extraction in ovarian cancer cell line co-cultures using deep neural networks.

Sharma O, Gudoityte G, Minozada R, Kallioniemi O, Turkki R, Paavolainen L Commun Biol. 2025; 8(1):303.

PMID: 40000764 PMC: 11862010. DOI: 10.1038/s42003-025-07766-w.

State of the ART: Drug Screening Reveals Artesunate as a Promising Anti-Fibrosis Therapy.

Qiao Y, Liang J, Jiang D J Respir Biol Transl Med. 2025; 2(1).

PMID: 39925974 PMC: 11800322. DOI: 10.70322/jrbtm.2024.10023.

Cell Painting for cytotoxicity and mode-of-action analysis in primary human hepatocytes.

Ewald J, Titterton K, Bauerle A, Beatson A, Boiko D, Cabrera A bioRxiv. 2025; .

PMID: 39896617 PMC: 11785178. DOI: 10.1101/2025.01.22.634152.

Distinguishing classes of neuroactive drugs based on computational physicochemical properties and experimental phenotypic profiling in planarians.

Ireland D, Rabeler C, Rao S, Richardson R, Collins E PLoS One. 2025; 20(1):e0315394.

PMID: 39883642 PMC: 11781733. DOI: 10.1371/journal.pone.0315394.

References

Gerry C, Hua B, Wawer M, Knowles J, Nelson Jr S, Verho O . Real-Time Biological Annotation of Synthetic Compounds. J Am Chem Soc. 2016; 138(28):8920-7. PMC: 4976700. DOI: 10.1021/jacs.6b04614. View

Warchal S, Unciti-Broceta A, Carragher N . Next-generation phenotypic screening. Future Med Chem. 2016; 8(11):1331-47. DOI: 10.4155/fmc-2016-0025. View

Gustafsdottir S, Ljosa V, Sokolnicki K, Wilson J, Walpita D, Kemp M . Multiplex cytological profiling assay to measure diverse cellular states. PLoS One. 2013; 8(12):e80999. PMC: 3847047. DOI: 10.1371/journal.pone.0080999. View

Subramanian A, Narayan R, Corsello S, Peck D, Natoli T, Lu X . A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles. Cell. 2017; 171(6):1437-1452.e17. PMC: 5990023. DOI: 10.1016/j.cell.2017.10.049. View

Yang K, Swanson K, Jin W, Coley C, Eiden P, Gao H . Analyzing Learned Molecular Representations for Property Prediction. J Chem Inf Model. 2019; 59(8):3370-3388. PMC: 6727618. DOI: 10.1021/acs.jcim.9b00237. View

Bemis G, Murcko M . The properties of known drugs. 1. Molecular frameworks. J Med Chem. 1996; 39(15):2887-93. DOI: 10.1021/jm9602928. View

Stokes J, Yang K, Swanson K, Jin W, Cubillos-Ruiz A, Donghia N . A Deep Learning Approach to Antibiotic Discovery. Cell. 2020; 180(4):688-702.e13. PMC: 8349178. DOI: 10.1016/j.cell.2020.01.021. View

Simm J, Klambauer G, Arany A, Steijaert M, Wegner J, Gustin E . Repurposing High-Throughput Image Assays Enables Biological Activity Prediction for Drug Discovery. Cell Chem Biol. 2018; 25(5):611-618.e3. PMC: 6031326. DOI: 10.1016/j.chembiol.2018.01.015. View

Wawer M, Jaramillo D, Dancik V, Fass D, Haggarty S, Shamji A . Automated Structure-Activity Relationship Mining: Connecting Chemical Structure to Biological Profiles. J Biomol Screen. 2014; 19(5):738-48. PMC: 5554950. DOI: 10.1177/1087057114530783. View

10.

Fernandez-Torras A, Comajuncosa-Creus A, Duran-Frigola M, Aloy P . Connecting chemistry and biology through molecular descriptors. Curr Opin Chem Biol. 2021; 66:102090. DOI: 10.1016/j.cbpa.2021.09.001. View

11.

Manica M, Oskooei A, Born J, Subramanian V, Saez-Rodriguez J, Rodriguez Martinez M . Toward Explainable Anticancer Compound Sensitivity Prediction via Multimodal Attention-Based Convolutional Encoders. Mol Pharm. 2019; 16(12):4797-4806. DOI: 10.1021/acs.molpharmaceut.9b00520. View

12.

Hofmarcher M, Rumetshofer E, Clevert D, Hochreiter S, Klambauer G . Accurate Prediction of Biological Assays with High-Throughput Microscopy Images and Convolutional Networks. J Chem Inf Model. 2019; 59(3):1163-1171. DOI: 10.1021/acs.jcim.8b00670. View

13.

Bray M, Singh S, Han H, Davis C, Borgeson B, Hartland C . Cell Painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes. Nat Protoc. 2016; 11(9):1757-74. PMC: 5223290. DOI: 10.1038/nprot.2016.105. View

14.

Dancik V, Carrel H, Bodycombe N, Seiler K, Fomina-Yadlin D, Kubicek S . Connecting Small Molecules with Similar Assay Performance Profiles Leads to New Biological Hypotheses. J Biomol Screen. 2014; 19(5):771-81. PMC: 5554958. DOI: 10.1177/1087057113520226. View

15.

Baell J, Holloway G . New substructure filters for removal of pan assay interference compounds (PAINS) from screening libraries and for their exclusion in bioassays. J Med Chem. 2010; 53(7):2719-40. DOI: 10.1021/jm901137j. View

16.

Way G, Zietz M, Rubinetti V, Himmelstein D, Greene C . Compressing gene expression data using multiple latent space dimensionalities learns complementary biological representations. Genome Biol. 2020; 21(1):109. PMC: 7212571. DOI: 10.1186/s13059-020-02021-3. View

17.

Chandrasekaran S, Ceulemans H, Boyd J, Carpenter A . Image-based profiling for drug discovery: due for a machine-learning upgrade?. Nat Rev Drug Discov. 2020; 20(2):145-159. PMC: 7754181. DOI: 10.1038/s41573-020-00117-w. View

18.

Caicedo J, McQuin C, Goodman A, Singh S, Carpenter A . Weakly Supervised Learning of Single-Cell Feature Embeddings. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2019; 2018:9309-9318. PMC: 6432648. DOI: 10.1109/CVPR.2018.00970. View

19.

Bray M, Gustafsdottir S, Rohban M, Singh S, Ljosa V, Sokolnicki K . A dataset of images and morphological profiles of 30 000 small-molecule treatments using the Cell Painting assay. Gigascience. 2017; 6(12):1-5. PMC: 5721342. DOI: 10.1093/gigascience/giw014. View

20.

Rohrer S, Baumann K . Maximum unbiased validation (MUV) data sets for virtual screening based on PubChem bioactivity data. J Chem Inf Model. 2009; 49(2):169-84. DOI: 10.1021/ci8002649. View