Robust AI-driven counter-measures: screening, guiding, combining

Abstract

Our goal is to develop an AI platform for drug discovery that supports rapid and cost-effective medical countermeasure (MC) development for emerging chemical and biological (CB) threats. This scenario commonly implies that we have limited knowledge About the new threat and that relevant screening data may be lacking. Therefore, traditional AI algorithms that require large, target-specific data for training are not directly applicable here. Moreover, we assume that the relevant bioactivity data will be collected iteratively, rather than available at the outset of the development program. The proposed in-silico screening tools must be helpful from the start in order to efficiently guide rapid response experimental efforts. Finally, these tools must be transparent to human experts, who ultimately lead drug design efforts and thus need to make decisions based on model predictions.

Document Details

Document Type
DoD Grant Award
Publication Date
Jun 14, 2022
Source ID
HDTRA12110013

Entities

People

  • Regina Barzilay

Organizations

  • Defense Threat Reduction Agency
  • Massachusetts Institute of Technology

Tags

Readers

  • Computational Modeling and Simulation
  • Distributed Systems and Data Platform Development
  • Geospatial Intelligence and Artificial Intelligence Analytics