About this seminar

Proteins and other biomolecules, such as DNA and RNA, are the minimal functional entities that realize life. Understanding how they execute their functions through their 3D structures and interaction dynamics provides a fundamental view of cellular life. This knowledge also allows us to exploit or modify these elegant molecules for a wide variety of purposes, such as drug design, immunotherapy, novel enzymes and others.

Since it is still challenging to experimentally study protein interactions at high-resolution, we combine the computational power of data-driven machine learning with physics-based molecular docking to better model 3D protein complexes. Specifically, we exploit deep learning and graph theory to tackle the major challenge in docking, namely scoring, the identification of correct conformations from a large pool of docked conformations, which still suffers from a low success rate. Both our graph-based approach (iScore) and our deep learning approach (DeepRank) show competitive performance compared with the state-of-the-art scoring functions. In addition, our DeepRank is a general, highly-configurable deep learning platform for data-mining 3D protein-protein interfaces. Both tools are freely available on Github.

Next, I plan to extend our AI-boosted 3D modelling framework to address the major puzzle in cancer vaccine design: which of the hundreds of a patient’s tumor mutations provide the best vaccine candidates to trigger the immune system into attacking tumors. I aim to reliably identify cancer vaccine candidates and model TCR:peptide:MHC structures, a key target protein complex for cancer vaccines. My Hypatia project strives to provide fundamental insights on how T cells respond to cancer vaccines and to improve the efficacy, safety and development time of personalized cancer vaccine design.

Key publications:
  • Geng, Cunliang, et al. "iScore: a novel graph kernel-based function for scoring protein–protein docking models." Bioinformatics 36.1 (2020): 112-121.
  • Geng, Cunliang, et al. "iSEE: Interface structure, evolution, and energy‐based machine learning predictor of binding affinity changes upon mutations." Proteins: Structure, Function, and Bioinformatics 87.2 (2019): 110-119.
  • DeepRank manuscript under preparation. Here is the Github link.
Agenda 18Feb2020 Seminar Li Xue

About this seminar

Bridging machine learning and physics-based docking for better modelling of biomolecular complexes.

read more

Practical information

  • Tuesday 18 February 2020 15:00 - 16:00 hrs.

Registration Not required


full professor