March 30, 2024Nature CommunicationsOpen Access

PLMSearch: Protein language model powers accurate and fast sequence search for remote homology

View Full Paper

Ask AI

Bookmark

Authors

WLWei LiuQB3 ZWZiye WangBeijing Information Science & Technology University RYRonghui YouNankai University

Discussion

Member takes

Implication

Computational evaluation demonstrates threefold higher sensitivity in remote homology detection across millions of protein pairs, suggesting sequence search can match structure-based accuracy.

Key Points

PLMSearch enables sensitive sequence search for remote homology by extracting deep contextual representations from a pre-trained protein language model.
Sensitivity increases by more than threefold over MMseqs2 while scanning millions of protein pairs in seconds, matching state-of-the-art structure search tools.
Computational benchmark on remote homology pairs reveals high recall of targets with dissimilar sequences but shared folds, which may enable fast protein annotation.

Cite This Study

Liu et al. (2024) studied this question.

synapsesocial.com/papers/68e719dbb6db643587693493 https://doi.org/10.1038/s41467-024-46808-5

View Full Paper

Ask AI

Bookmark

Implication

Computational evaluation demonstrates threefold higher sensitivity in remote homology detection across millions of protein pairs, suggesting sequence search can match structure-based accuracy.

PLMSearch: Protein language model powers accurate and fast sequence search for remote homology

Discussion

Member takes

Implication

Key Points

Cite This Study

PLMSearch: Protein language model powers accurate and fast sequence search for remote homology

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider