How a researcher uses Codex and ChatGPT to search for new antimicrobial molecules

César de la Fuente’s cross-disciplinary lab is utilizing OpenAI's Codex and ChatGPT alongside deep-learning models to search living and extinct genomes for antimicrobial candidates. With drug-resistant microbes posing a major global health threat and approximately five million deaths associated with bacterial antimicrobial resistance in 2021, traditional drug discovery methods offer diminishing returns. By treating biology as an information system, the lab's models recognize patterns in biological sequences to reduce the initial search for candidate molecules from years to hours. In addition to deep-learning models, the lab uses ChatGPT and Codex to brainstorm hypotheses, write and refine code, process large datasets, analyze results, and connect ideas across scientific disciplines. Because the lab brings together experts from biology, chemistry, computer science, and engineering, the AI tools help bridge gaps between fields, allowing biologists to build programs and programmers to tackle biological problems. Lab members also use ChatGPT to review unfamiliar topics, clarify terminology, compare methods, and work in their native languages. While de la Fuente values AI as a brainstorming partner and collaborative sounding board, he cautions that researchers must always double-check the outputs for accuracy. Ultimately, de la Fuente emphasizes that AI predictions must be validated by ground-truth laboratory experiments to ensure candidate molecules are effective, safe, and viable for clinical trials.