Mengchen Pu

This study reveals that Arpin homodimerization enhances inhibition of Arp2/3 complex-mediated actin polymerization, suppressing cancer cell migration and metastasis.

The paper introduces OriginFlow, a protein generative model that achieves high success rates in designability, diversity, and functional design of proteins, validated through wet-lab experiments.

This paper introduces SLWise, a deep-learning model for predicting synthetic lethal interactions in specific cellular contexts, and SLAD-CE for analyzing their mechanisms, showing improved accuracy over existing methods.

This study uses a deep learning model to identify mutational signatures in metastatic breast cancer and link them to therapeutic targets.

The paper introduces Cosinet, a tool for single-sample differential co-expression analysis in breast cancer, showing it can predict clinical outcomes like survival and recurrence.

This paper introduces a deep learning model using graph-based representation to predict cell-specific synthetic lethal gene pairs, enhancing cancer precision medicine.

The paper introduces LLPSWise, a method for predicting LLPS constituents and analyzing the human proteome to provide a systematic view of LLPS systems.