I-Corps: Translation potential of an artificial intelligence (AI) platform that combines systems biology with structural predictions for small molecule drug discovery

This I-Corps project is based on the development of an artificial intelligence (AI) drug development process to generate drugs specific to various diseases. Current approaches to drug discovery rely on costly screening or simulations that fail to capture the complex biological context in which drugs operate. These methods typically examine proteins in isolation rather than within their biological pathways, leading to high failure rates due to unforeseen off-target effects and poor efficacy. This technology addresses these challenges by integrating a systems-level understanding with structural biology through the Bone Morphogenetic Protein (BMP) pathway foundation model. BMPs are a group of proteins that play a crucial role in bone and cartilage formation, as well as in various other developmental processes. This approach may predict how drugs interact within complex biological networks rather than just with individual targets. The solution may create more accurate predictions of both therapeutic potential and possible side effects, potentially improving the drug discovery process and patient survival rates while lowering the average cost of drug development.

This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of an integrated AI platform that combines systems biology with structural predictions for more effective small molecule drug discovery. This technology is based on a therapeutic generative diffusion model that proposes novel drugs and a transformer-based foundation model of systems biology to prioritize biological targets of interest. Current structure-based drug discovery systems often optimize small molecules according to a target of interest without considering off-target effects. The technology unites a novel systems biology foundation model for the Bone Morphogenetic Protein (BMP) pathway with a guided diffusion model for small molecule generation. This integration may enable the targeting of key protein interactions while minimizing common off-target effects like nausea or hair loss during chemotherapy. The key advance over existing solutions is the inclusion of systems biology data in the generation of novel therapeutics to reduce the probability of off-target effects, which are often responsible for the failure of clinical trials. This solution may allow pharmaceutical companies to repurpose existing drugs using knowledge from this foundation model, generate alternative therapeutics for existing targets to hedge development programs, prioritize experiments to reduce the time to market, and decrease failure rates in clinical trials.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2534280
Principal Investigator: Elliot Hui
Funds Obligated: $50,000
State: CA