A Bayesian active learning platform for scalable combination drug screens Journal Article
| Authors: | Tosh, C.; Tec, M.; White, J. B.; Quinn, J. F.; Ibanez Sanchez, G.; Calder, P.; Kung, A. L.; Dela Cruz, F. S.; Tansey, W. |
| Article Title: | A Bayesian active learning platform for scalable combination drug screens |
| Abstract: | Large-scale combination drug screens are generally considered intractable due to the immense number of possible combinations. Existing approaches use ad hoc fixed experimental designs then train machine learning models to impute unobserved combinations. Here we propose BATCHIE, an orthogonal approach that conducts experiments dynamically in batches. BATCHIE uses information theory and probabilistic modeling to design each batch to be maximally informative based on the results of previous experiments. On retrospective experiments from previous large-scale screens, BATCHIE designs rapidly discover highly effective and synergistic combinations. In a prospective combination screen of a library of 206 drugs on a collection of pediatric cancer cell lines, the BATCHIE model accurately predicts unseen combinations and detects synergies after exploring only 4% of the 1.4M possible experiments. Further, the model identifies a panel of top combinations for Ewing sarcomas, which follow-up validation experiments confirm to be effective, including the rational and translatable top hit of PARP plus topoisomerase I inhibition. These results demonstrate that adaptive experiments can enable large-scale unbiased combination drug screens with a relatively small number of experiments. BATCHIE is open source and publicly available (https://github.com/tansey-lab/batchie). © The Author(s) 2024. |
| Keywords: | osteosarcoma; controlled study; human cell; aurora kinase inhibitor; area under the curve; dose response; drug potentiation; monotherapy; nonhuman; antineoplastic agents; cytarabine; temozolomide; topotecan; follow up; antineoplastic agent; prospective study; quality control; animal cell; mouse; dna damage; bayesian learning; bayes theorem; gene expression; antineoplastic combined chemotherapy protocols; in vivo study; high throughput screening; in vitro study; drug screening; pathology; drug screening assays, antitumor; enzyme activity; validation study; cell line, tumor; retrospective study; prediction; irinotecan; ewing sarcoma; sarcoma; drug synergism; simulation; tumor cell line; drug response; tipifarnib; epirubicin; nicotinamide adenine dinucleotide adenosine diphosphate ribosyltransferase inhibitor; tumor model; dna topoisomerase inhibitor; rhabdomyosarcoma; drug therapy; mitomycin; anthracycline; ex vivo study; information science; experimental design; clofarabine; performance; complementary dna; dna topoisomerase; cladribine; aurora a kinase; small cell lung cancer; experimental study; cytotoxicity assay; drug; procedures; machine learning; learning algorithm; detection method; entropy; therapeutic index; trametinib; deoxyadenosine; sarcoma, ewing; predictive model; alisertib; cell component; cancer; humans; human; article; cancer cell line; ic50; clustered regularly interspaced short palindromic repeat; generative model; talazoparib; eltanexor; sjsa-1 cell line |
| Journal Title: | Nature Communications |
| Volume: | 16 |
| ISSN: | 2041-1723 |
| Publisher: | Nature Publishing Group |
| Date Published: | 2025-01-02 |
| Start Page: | 156 |
| Language: | English |
| DOI: | 10.1038/s41467-024-55287-7 |
| PUBMED: | 39746987 |
| PROVIDER: | scopus |
| PMCID: | PMC11696745 |
| DOI/URL: | |
| Notes: | Article -- MSK Cancer Center Support Grant (P30 CA008748) acknowledged in PubMed and PDF -- MSK corresponding author is Wesley Tansey -- Source: Scopus |
Altmetric
Citation Impact
BMJ Impact Analytics
Related MSK Work