Qianmu Yuan

This paper introduces InterAb and InterAb-Opt, computational frameworks that predict and optimize broadly neutralizing antibodies using all-atom modeling and pre-trained language models, showing significant improvements in binding affinity and neutralization activity.

ProTact improves protein-protein interaction prediction using a geometric graph neural network with physics-informed constraints, outperforming existing methods on benchmark datasets.

MAProt improves protein design by combining structure-based and language model approaches using multi-agent reinforcement learning with Pareto-based optimization, achieving better performance across multiple tasks.

GPSite predicts protein binding sites across multiple ligands using a multi-task network, outperforming existing methods and enabling genome-scale annotations.

GLMSite identifies DNA and RNA-binding sites using geometric graph learning on ESMFold-predicted structures, outperforming sequence-based methods and matching structure-based approaches.

GraphBepi predicts B-cell epitopes using AlphaFold2 structures and a pretrained language model, outperforming existing methods in AUC and AUPR.

SPROF-GO predicts protein functions from sequences using a pretrained language model and label diffusion, outperforming existing methods by over 10% in AUPR.

LMetalSite predicts metal ion-binding sites from protein sequences using a pretrained language model and multi-task learning, achieving higher accuracy than structure-based methods.

The paper introduces GraphSite, a method using AlphaFold2-predicted structures and graph transformers to predict DNA-binding residues, outperforming existing sequence and structure-based methods.