June 13
Li to Discuss Transformative AI
NIH’s Office of Data Science Strategy hosts a seminar series to highlight models of data sharing and reuse on the second Friday of each month. The next seminar, “Transformative AI for Deep Mining of Omics and Literature Data,” will take place on June 13 at noon.
Transformative artificial intelligence (AI) models are powerful tools for large-scale mining of biomedical data. In this talk, Dr. Fuhai Li will present novel approaches his lab has developed to combine large language models (LLMs) with graph-based AI to integrate and analyze vast omics datasets for identifying disease targets, mapping signaling pathways and predicting effective drug combinations. The key component of this novel AI system is the text-numeric graph (TNG), a structure in which graph entities and associations carry both textual and numeric attributes.
Li, associate professor in Washington University’s school of medicine and computer science & engineering, will also introduce an AI multi-agent system that his team developed to accelerate biomedical discovery by unifying omics data analysis, literature-based deep search and reasoning to generate novel scientific hypotheses. He will also showcase the applications of these novel AI tools with analysis of heterogeneous pharmacogenomics data for cancer research.
Before joining WU, Li was an assistant professor at Ohio State University. He received his Ph.D. in applied mathematics in Beijing University and completed his postdoc training at Harvard Medical School in computational biology.
To register for this seminar, see Data Science June lecture.