Dr. Li Xue’s research interests focus on bridging artificial intelligence (AI) and structural biology to better understand the molecular basis of diseases and to rationalize drug design. Dr. Xue is a multidisciplinary researcher with strong backgrounds in artificial intelligence (MS, PhD) and structural biology (VENI fellow in the HADDOCK group, the flagship software for protein docking).In the past five years, Dr. Xue initiated an independent research line: AI-boosted 3D modelling (VENI and eScience grants). 3D modelling of protein complexes faces two major challenges: 1) sampling: protein flexibility entails a huge sampling space thus heavy computational loads, and 2) scoring: our understanding of how proteins interact are still not enough to accurately distinguish native-like conformations from a huge pool of models. To tackle these challenges, Dr. Xue focuses on exploiting interaction patterns learned by AI from the huge amount of existing protein complexes for 1) better and faster 3D modelling and 2) predicting mutation effects on binding affinities of protein complexes (e.g., iScore and iSEE papers). Dr. Xue further established a deep learning framework for data-mining 3D protein complex images, DeepRank (https://github.com/DeepRank
), in collaboration with the Netherlands eScience Center and SurfSara.Next, Dr. Xue plans to work closely with cancer immunotherapy groups in Radboudumc to decipher TCR specificities and aid cancer immunotherapy design.