Nearest neighbor search on embeddings rapidly identifies distant protein relations
Open Access
- 17 November 2022
- journal article
- research article
- Published by Frontiers Media SA in Frontiers in Bioinformatics
- Vol. 2, 1033775
Abstract
Since 1992, all state-of-the-art methods for fast and sensitive identification of evolutionary, structural, and functional relations between proteins (also referred to as “homology detection”) use sequences and sequence-profiles (PSSMs). Protein Language Models (pLMs) generalize sequences, possibly capturing the same constraints as PSSMs, e.g., through embeddings. Here, we explored how to use such embeddings for nearest neighbor searches to identify relations between protein pairs with diverged sequences (remote homology detection for levels of <20% pairwise sequence identity, PIDE). While this approach excelled for proteins with single domains, we demonstrated the current challenges applying this to multi-domain proteins and presented some ideas how to overcome existing limitations, in principle. We observed that sufficiently challenging data set separations were crucial to provide deeply relevant insights into the behavior of nearest neighbor search when applied to the protein embedding space, and made all our methods readily available for others.Keywords
All Related Versions
This publication has 66 references indexed in Scilit:
- HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignmentNature Methods, 2011
- HMMER web server: interactive sequence similarity searchingNucleic Acids Research, 2011
- Hidden Markov model speed heuristic and iterative HMM search procedureBMC Bioinformatics, 2010
- Comprehensive Evaluation of Protein Structure Alignment Methods: Scoring by Geometric MeasuresJournal of Molecular Biology, 2005
- Enzyme Function Less Conserved than AnticipatedJournal of Molecular Biology, 2002
- Within the twilight zone: a sensitive profile-profile comparison tool based on information theoryJournal of Molecular Biology, 2002
- Gapped BLAST and PSI-BLAST: a new generation of protein database search programsNucleic Acids Research, 1997
- CATH – a hierarchic classification of protein domain structuresStructure, 1997
- Hidden Markov Models in Computational BiologyJournal of Molecular Biology, 1994
- Identification of common molecular subsequencesJournal of Molecular Biology, 1981