Speaker
Description
Predicting interactions between proteins is fundamental since protein–protein complexes are crucial in physiological and pathological processes. Despite this, understanding the binding process is challenging: binding partners must find each other amid thousands of other molecules in the crowded cell. This binding specificity is determined by a complex combination of matches at the interfaces. Finding a balance between accuracy and efficiency when studying these many features is a difficult task. From this, the importance of developing computationally efficient methods.
We presented an integrated protocol based on compact modeling of protein surface patches via sets of orthogonal polynomials to identify regions of high shape/electrostatic complementarity [1]. By incorporating both hydrophilic and hydrophobic contributions, we defined new binding matrices to train a neural network. Our architecture, the Core Interacting Residues Network (CIRNet), achieved a performance in terms ROC AUC of ~0.87 in identifying pairs of core interacting residues on a balanced data set.
We tested this protocol to enhance docking algorithms by filtering the proposed poses, addressing one of the still open problems in computational biology. When applied to the top ten models from three widely used docking servers, CIRNet improved docking outcomes, significantly reducing the average RMSD between the selected poses and the native state. Compared to another state-of-the-art tool for rescaling docking poses, CIRNet more efficiently identified the worst poses generated by the three docking servers under consideration and achieved superior rescaling performance in two cases.
[1] Grassmann, G, et al. JCIM (2024).
| Role | Post Doc |
|---|
