ArticleNature communications2022
Protein language models trained on multiple sequence alignments learn phylogenetic relationships.
Article in Nature communications, 2022. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 35 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
35 citing papers in PubMed.
- Hierarchical Organization of Evolutionary Constraint in Duplicated EF-Hand Motifs.Journal of molecular evolution · 2026Article
- A spectral framework for measuring diversity in multiple sequence alignments.PLoS computational biology · 2026Article
- Analysis of phylogenetic signal in protein language model embeddings.Scientific reports · 2026Article
- Protein language models are accidental taxonomists.BMC bioinformatics · 2026Article
- ProteomeLM: A proteome-scale language model enables accurate and rapid prediction of protein-protein interactions and gene essentiality across taxa.Proceedings of the National Academy of Sciences of the United States of America · 2026Article
- Repurposing AI for protein interactions and dynamics: opportunities, limitations, and lessons.Frontiers in bioinformatics · 2026Review
- Deep learning on protein language model embeddings unlocks accurate prediction of protein solubility.Frontiers in microbiology · 2026Article
- AI-driven insights into protein misfolding and innate immunity in neurodegenerative diseases.Frontiers in immunology · 2026Review
- Phylogenetic corrections and higher-order sequence statistics in protein families: Potts vs multiple sequence alignment transformer machine learning models.Physical review research · 2025Article
- OneProt: Towards multi-modal protein foundation models via latent space alignment of sequence, structure, binding sites and text encoders.PLoS computational biology · 2025Article
- Protein Language Models are Accidental Taxonomists.bioRxiv : the preprint server for biology · 2025Article
- Phylogenetic Methods Meet Deep Learning.Genome biology and evolution · 2025Article
- Protein Structural Phylogenetics.Genome biology and evolution · 2025Review
- Artificial intelligence-powered prediction of AIM-2 inflammasome sequences using transformers and graph attention networks in periodontal inflammation.Scientific reports · 2025Article
- Major advances in protein function assignment by remote homolog detection with protein language models - A review.Current opinion in structural biology · 2025Review
- Advances of deep Neural Networks (DNNs) in the development of peptide drugs.Future medicinal chemistry · 2025Review
- CATHe2: Enhanced CATH superfamily detection using ProstT5 and structural alphabets.Biology methods & protocols · 2025Article
- Fine-tuning protein language models unlocks the potential of underrepresented viral proteomes.PeerJ · 2025Article
- General Intelligence Framework to Predict Virus Adaptation Based on a Genome Language Model.Research (Washington, D.C.) · 2025Article
- G-Computational and structural biotechnology journal · 2024Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
3 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
Self-supervised neural language models with attention have recently been applied to biological sequence data, advancing structure, function and mutational effect prediction. Some protein language models, including MSA Transformer and AlphaFold's EvoFormer, take multiple sequence alignments (MSAs) of evolutionarily related proteins as inputs. Simple combinations of MSA Transformer's row attentions have led to state-of-the-art unsupervised structural contact prediction. We demonstrate that similarly simple, and universal, combinations of MSA Transformer's column attentions strongly correlate with Hamming distances between sequences in MSAs. Therefore, MSA-based language models encode detailed phylogenetic relationships. We further show that these models can separate coevolutionary signals encoding functional and structural constraints from phylogenetic correlations reflecting historical contingency. To assess this, we generate synthetic MSAs, either without or with phylogeny, from Potts models trained on natural MSAs. We find that unsupervised contact prediction is substantially more resilient to phylogenetic noise when using MSA Transformer versus inferred Potts models.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.