Lee Jin Sub, Kim Philip M
Department of Molecular Genetics, University of Toronto, Toronto, Ontario, M5S 3K3, Canada.
Department of Computer Science, University of Toronto, Toronto, Ontario, M5S 2E4, Canada.
Bioinformatics. 2025 Mar 4;41(3). doi: 10.1093/bioinformatics/btaf010.
Accurate prediction of protein side-chain conformations is necessary to understand protein folding, protein-protein interactions and facilitate de novo protein design.
Here, we apply torsional flow matching and equivariant graph attention to develop FlowPacker, a fast and performant model to predict protein side-chain conformations conditioned on the protein sequence and backbone. We show that FlowPacker outperforms previous state-of-the-art baselines across most metrics with improved runtime. We further show that FlowPacker can be used to inpaint missing side-chain coordinates and also for multimeric targets, and exhibits strong performance on a test set of antibody-antigen complexes.
Code is available at https://gitlab.com/mjslee0921/flowpacker.
准确预测蛋白质侧链构象对于理解蛋白质折叠、蛋白质-蛋白质相互作用以及促进从头蛋白质设计至关重要。
在此，我们应用扭转流匹配和等变图注意力来开发FlowPacker，这是一种快速且性能良好的模型，用于根据蛋白质序列和主链预测蛋白质侧链构象。我们表明，FlowPacker在大多数指标上优于先前的最先进基线，且运行时间有所改善。我们进一步表明，FlowPacker可用于修复缺失的侧链坐标，也可用于多聚体目标，并在抗体-抗原复合物测试集上表现出强大性能。
