Dr. Li Xue: Utrecht University, Netherlands, Assistant Professor of Chemistry
Title: Artificial intelligence boosted 3D modelling of protein-protein complexes and the estimation of binding affinity changes upon mutations
Abstract: Quantitative evaluation of binding affinity change upon mutations is crucial for protein engineering and drug design. Machine learning based methods are gaining increasing momentum in this field. Here, we propose a fast and reliable predictor of binding affinity changes upon single point mutation effects based on a random forest approach. Our method, iSEE, uses a limited number of interface, Structure, Evolution and Energy-based features. Due to the limited number of experimental data, using a small number of sensitive predictive features is vital to the generalization and robustness of such machine learning methods. iSEE achieves, using only 31 features, a high Pearson correlation coefficient of 0.80 and a root mean square error of 1.4 kcal mol-1 on a diverse dataset consisting of 1102 mutations in 57 protein-protein complexes. It outperforms existing state-of-the-art methods on two blind test datasets. Feature analysis underlines the significance of evolutionary conservations for quantitative prediction of mutation effects. As an application example, we perform a full mutation scanning of the interface residues in the MDM2-p53 complex.
Host: Drena Dobbs
Biography: Dr. Xue is an ISU alumna. She received her PhD in BCB (2012) with Vasant Honavar, Computer Science, and Drena Dobbs, GDCB, recently completed a postdoc in the Netherlands with Alex Bonvin (developer of HADDOCK docking server), where she received a highly competitive and prestigious Veni Postdoctoral Fellowship for her work in the Bijvoet Center for Biomolecular Research at Utrecht.
She is currently in the Computational Structural Biology Group at Bijvoet Center for Biomolecular Research on the Faculty of Science for Chemistry at Utrecht University in the Netherlands.
Below are URLs that include links to her CV and some recent publications, also her Google Scholar page:
- Performance of HADDOCK and a simple contact-based protein-ligand binding affinity predictor in the D3R Grand Challenge 2.
- iSEE: Interface Structure, Evolution and Energy-based random forest predictor of binding affinity changes upon mutations. C. Geng, A. Vangone, G.E. Folkers, L. Xue* and Alexandre M.J.J. Bonvin*
- Kurkcuoglu Z, Koukos PI, Citro N, Trellet ME, Rodrigues JPGLM, Moreira IS, Roel-Touris J, Melquiond ASJ, Geng C, Schaarschmidt J, , Vangone A, Bonvin AMJJ. . 2017 Aug 22. doi: 10.1007/s10822-017-0049-y. [Epub ahead of print]