Reproduce “CoMoFlow: Controllable Protein Conformational Ensemble Generation via Explicit Collective-Mode Modeling” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Yusong Wang, Zheyuan Gu, Ailiang Lin, Minghao Shao, Yicheng Xu, Jiang Cai, Prayag Tiwari, Mingkun Xu

The authors introduce CoMoFlow, which generates protein backbone ensembles by learning protein-specific collective modes and guiding an SE(3) flow with sampled mode coefficients. On ATLAS, compared with BBFlow, it raises DCCM correlation from 0.87 to 0.93 and lowers PCA W2 from 1.33 to 1.21. These results concern matching MD-derived backbone ensembles, not unobserved states or biological function.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.