Hypothesis Generation with AGATHA: Accelerate Scientific Discovery with Deep Learning

Ilya Tyagin

PhD Student in Bioinformatics Data Science

September 26, 2022 | 3:30 PM

Ammon-Pinizzotto Biopharmaceutical Innovation (BPI) Building
Conference Room 140

Medical research is risky and expensive. Drug discovery requires researchers to efficiently winnow thousands of potential targets to a small candidate set. However, scientists spend significant time and money long before seeing the intermediate results that ultimately determine this smaller set. Hypothesis generation systems address this challenge by mining the wealth of publicly available scientific information to predict plausible research directions. We present AGATHA, a deep-learning hypothesis generation system that learns a data-driven ranking criteria to recommend new biomedical connections. We massively validate our system and explore biomedical sub-domains demonstrating AGATHA’s predictive capacity across the most popular relationship types. Overall, AGATHA achieves high recommendation quality when compared to other hypothesis generation systems built to predict across all available biomedical literature.

Ilya Tyagin is a PhD student in the Bioinformatics Data Science program at UD. He is doing research in biomedical hypothesis generation in the Algorithms, AI and Computational Science lab under the supervision of Dr. Ilya Safro. Ilya Tyagin got his Bachelor degree in Applied Mathematics and Physics with a minor in Bioinformatics from Moscow Institute of Physics and Technology, and conducted his bachelor thesis at Vavilov Institute of General Genetics. His research interests include Deep learning, High-Performance Computing, Natural Language Processing and Literature-Based discovery.

Download Flyer