Kallenborn Felix, Chacon Alejandro, Hundt Christian, Sirelkhatim Hassan, Didi Kieran, Cha Sooyoung, Dallago Christian, Mirdita Milot, Schmidt Bertil, Steinegger Martin
Department of Computer Science, Johannes Gutenberg University Mainz, Mainz, Germany.
NVIDIA, Santa Clara, CA, USA.
Nat Methods. 2025 Sep 18. doi: 10.1038/s41592-025-02819-8.
Rapidly growing protein databases demand faster sensitive search tools. Here the graphics processing unit (GPU)-accelerated MMseqs2 delivers 6× faster single-protein searches than CPU methods on 2 × 64 cores, speeds previously requiring large protein batches. For larger query batches, it is the most cost-effective solution, outperforming the fastest alternative method by 2.4-fold with eight GPUs. It accelerates protein structure prediction with ColabFold 31.8× over the standard AlphaFold2 pipeline and protein structure search with Foldseek by 4-27×. MMseqs2-GPU is available under an open-source license at https://mmseqs.com/ .
快速增长的蛋白质数据库需要更快且灵敏的搜索工具。在此，图形处理单元（GPU）加速的MMseqs2在2×64核的配置下，单蛋白搜索速度比CPU方法快6倍，而这样的速度此前需要处理大量蛋白质批次才能实现。对于更大的查询批次，它是最具成本效益的解决方案，在使用八个GPU时比最快的替代方法性能高出2.4倍。它能使ColabFold的蛋白质结构预测速度比标准AlphaFold2流程快31.8倍，使Foldseek的蛋白质结构搜索速度快4至27倍。MMseqs2-GPU可在https://mmseqs.com/ 以开源许可获取。
