3 Questions: Using AI to accelerate the discovery and design of therapeutic drugs

2026-02-04 · Source: MIT News - Artificial intelligence · Field: Science & Research — Health & Medical Research, Mathematics & Computational Sciences · Depth: Advanced, medium

Summary

Professor James Collins, a co-founder of synthetic biology and director of the MIT Jameel Clinic for Machine Learning in Health, discusses his research into using AI for therapeutic drug discovery and design. His work, often in collaboration with institutions like the Broad Institute and Wyss Institute, combines computational predictions with advanced experimental platforms. Key advances include the 2020 discovery of halicin, a potent antibiotic effective against multidrug-resistant bacteria, through deep learning. In 2025, his lab demonstrated using generative AI to design novel antibiotics from scratch, leading to compounds like NG1, which targets *Neisseria gonorrhoeae*, and DN1, effective against MRSA. Collins also co-founded Phare Bio, a nonprofit aiming to advance AI-discovered antibiotic candidates toward clinical development, recently securing an ARPA-H grant to design 15 new preclinical antibiotics.

Key takeaway

For AI Researchers and drug developers focused on antimicrobial resistance, your teams should prioritize integrating generative AI with high-throughput biological testing to accelerate the discovery and design of novel, safe, and effective antibiotics. Consider forming interdisciplinary collaborations, like those at the MIT Jameel Clinic, to combine AI, network biology, and systems microbiology expertise, moving from reactive to proactive antibiotic development strategies.

Key insights

AI, particularly deep learning and generative models, significantly accelerates novel antibiotic discovery and design.

Principles

• Collaboration is central to combining diverse expertise.
• Integrate computational predictions with experimental platforms.
• Generative AI can design entirely new molecules.

Method

Deep learning and generative AI (genetic algorithms, variational autoencoders) are used to generate candidate molecules, followed by computational filtering, retrosynthetic modeling, medicinal chemistry review, and experimental testing.

In practice

• Use organs-on-chips to test drug efficacy in human-like environments.
• Explore fragment-based and unconstrained chemical space for drug design.
• Integrate AI with high-throughput biological testing.

Topics

• AI in Drug Discovery
• Antibiotic Discovery
• Generative AI
• Deep Learning
• Synthetic Biology

Best for: AI Researcher, AI Scientist, Research Scientist

Related on AIssential

• The scientist using AI to hunt for antibiotics just about everywhere — AI-driven peptide discovery from diverse biological sources offers a novel approach to combat antimicrobial resistance.
• Using AI to Develop New Drugs — AI is fundamentally transforming drug discovery and development processes.
• Using synthetic biology and AI to address global antimicrobial resistance threat — Synthetic biology and AI can develop programmable antibacterials to combat rising antimicrobial resistance.
• Using AI to outsmart drug-resistant bacteria — AI tools significantly accelerate antibiotic discovery by identifying non-intuitive biological patterns and generating novel therapeutic ideas.
• Data-driven surrogates of rational design enable antimicrobial peptide optimization — Generative AI and data-driven surrogates enable rapid optimization of antimicrobial peptides against drug resistance.
• NVIDIA & Eli Lilly: The AI Revolution in Drug Discovery | Jensen Huang & David Ricks — Co-designing computing stacks for specific problems accelerates applications far beyond Moore's Law.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MIT News - Artificial intelligence.