Development of a metabolomics enabled AI/ML platform for discovery of new treatments to enhance drug sensitivity in cancer

Project Summary Coupling high-throughput omics technologies with machine learning and the phenotypic-based drug discovery paradigm allows for data-driven drug discovery (D4). D4 has the advantage of being unbiased, like phenotypic- based drug discovery, but the comprehensive measurements of 100s to 100,000s of biological features also enables characterizing drug perturbations and disease signatures to gain mechanistic insights. The predominant data type for D4 has been transcriptomics and more recently image-based assays. Other biomolecules (i.e., proteins and metabolites) are considered closer to the phenotype, however technical challenges in data generation and analysis, as well as the lack of standardized data analysis pipelines have limited the systematic use of these data types. Sinopia Bioscience and Omix Technologies have combined their strengths in systems biology data analysis, AI/ML, and LC-MS/MS based metabolomics to develop a unique metabolomics-based platform that has allowed for systematic metabolic characterization of a chemical library consisting of ~3,300 small molecules covering more than 1,000 drug targets. Our preliminary results demonstrate that metabolomics is more sensitive, reproducible, and predictive of the mechanism of action of these small molecules than transcriptomics. Further, we found that using metabolomics data we could predict cell line specific toxicity of cancer drugs in viability assays. Importantly, we found we could derive metabolic signatures of sensitivity and resistance and use these to identify secondary compounds that can enhance sensitivity to cancer drugs. In this Phase I proposal, we will expand on these findings and develop novel algorithms to better understand how baseline metabolic states of cancers affect their sensitivity to cancer drugs. As a development test case we will focus on sensitivity of breast cancer cell lines to docetaxel and tucatinib. First, we will perform high throughput metabolomics and analysis of 100 cancer cell lines to characterize metabolism both in a baseline state and after administration of docetaxel and tucatinib. Second, we will develop novel computational algorithms for predicting the ability of compounds to sensitize cancer cells to cancer drugs. Finally, we will experimentally validate novel compound combinations and generate metabolomics data to further improve our algorithms. Success of this Phase I proposal will lead to validated metabolomics-based methods for identifying underlying metabolic phenotypes predictive of drug sensitivity, which will then be leveraged to predict the effects of compound combinations. This allows us to further expand Sinopia’s platform’s capabilities and its application to oncology applications through partnerships with biotech/pharma and/or fundraising through outside investors. In addition, it would lead to novel use of targets and compounds to enhance the sensitivity to existing cancer treatments that we can internally develop. Phase II will focus on further development of the platform, expanding our metabolomics-based library, and advancing promising synergistic combinations into preclinical models.