Abstract B031: OneThree Biotech’s artificial intelligence platform for optimizing drug development
Coryandar M. Gilvary, Neel S. Madhukar, Olivier Elemento · Molecular Cancer Therapeutics · 2019
Abstract Recently, it has been shown how artificial intelligence (AI) has the possibility to dramatically shorten the drug development pipeline by identifying insights that may have otherwise been missed. However, one major critique of these methods has been their black-box nature and the lack of mechanistic biology. To address these issues, OneThree Biotech has developed an extensive platform of biology-driven AI approaches that accelerate early stage drug development. Beginning with novel target identification, we present ECLIPSE, an AI approach that combines genomic, cell line and experimental design features to identify essential genes within cancer cell lines, based upon CRISPR and shRNA loss-of-function screenings. We demonstrated that ECLIPSE could accurately identify known and potential cancer targets, as well as be used to determine drug efficacy in the clinic. In cases where a compound’s target is unknown, we introduce BANDIT, a Bayesian model that combines divergent data sources to predict the targets and mechanisms for small molecules with unprecedented accuracy and versatility. Using BANDIT, we successfully identified a novel class of microtubule inhibitors and a previously unknown mechanism of ONC201, an anti-cancer small molecule, which led to a successful Phase 2 trial in a rare glioblastoma. Building on these approaches, we also introduce our suite of drug synergy prediction models. These models can not only predict the level of expected drug synergy in specific cancer types, but can also be used to pinpoint the specific mechanisms of actions that contribute to synergy. Altogether, OneThree’s platform is a comprehensive AI approach that combines high quality AI with mechanistic biology to optimize early-stage drug development. Citation Format: Coryandar Gilvary, Neel Madhukar, Olivier Elemento. OneThree Biotech’s artificial intelligence platform for optimizing drug development [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2019 Oct 26-30; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2019;18(12 Suppl):Abstract nr B031. doi:10.1158/1535-7163.TARG-19-B031