SigSpace: an LLM-based agent for drug response signature interpretation

Rohit Khurana, Ishita Mangla, Giovanni Palla, Siddhant Sanghi, Daniele Merico

Published: 23 Sept 2025, Last Modified: 26 Sept 2025AI4D3 2025 PosterEveryoneRevisionsBibTeXCC BY 4.0
Keywords: computational biology, drug discovery, agent systems, LLMs
TL;DR: This study demonstrates that LLM-powered agents can effectively interpret and summarize transcriptional responses of cancer cells to drug treatments, outperforming random baselines in matching drug signatures to their mechanisms of action.
Submission Number: 21