Protein Protein Interaction
Protein-protein interactions (PPIs) are fundamental to virtually all biological processes, and accurately predicting them is crucial for understanding disease mechanisms and developing new therapeutics. Current research heavily utilizes machine learning, employing diverse architectures such as graph neural networks, transformers, and large language models to analyze protein sequences and structures, often incorporating multi-view learning and pre-training strategies to improve prediction accuracy and generalizability. These advancements are significantly impacting drug discovery and protein engineering by enabling more efficient identification of drug targets and the design of novel protein-based therapies.
Papers
Revealing data leakage in protein interaction benchmarks
Anton Bushuiev, Roman Bushuiev, Jiri Sedlar, Tomas Pluskal, Jiri Damborsky, Stanislav Mazurenko, Josef Sivic
Graph Neural Networks for Protein-Protein Interactions -- A Short Survey
Mingda Xu, Peisheng Qian, Ziyuan Zhao, Zeng Zeng, Jianguo Chen, Weide Liu, Xulei Yang