The DOI matches its Crossref record; 66 of the 66 references it cites were found in a registry or an index.
Using deep learning to annotate the protein universe
Open accessPublished version of a preprint
Abstract
Understanding the relationship between amino acid sequence and protein function is a long-standing challenge with far-reaching scientific and translational implications. State-of-the-art alignment-based techniques cannot predict function for one-third of microbial protein sequences, hampering our ability to exploit data from diverse organisms. Here, we train deep learning models to accurately predict functional annotations for unaligned amino acid sequences across rigorous benchmark assessments built from the 17,929 families of the protein families database Pfam. The models infer known patterns of evolutionary substitutions and learn representations that accurately cluster sequences from unseen families. Combining deep models with existing methods significantly improves remote homology detection, suggesting that the deep models learn complementary information. This approach extends the coverage of Pfam by >9.5%, exceeding additions made over the last decade, and predicts function for 360 human reference proteome proteins with no previous Pfam annotation. These results suggest that deep learning models will be a core component of future protein annotation tools.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“These embeddings can capture evolutionary traits of the proteins, as well as functional and structural characteristics.119”
From Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence (2023) · cites this paper as [119]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Basic local alignment search tool
Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990) · Journal of Molecular Biology · doi:10.1006/jmbi.1990.9999 · Cited by 98,669
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs
Altschul, S. (1997) · Nucleic Acids Research · doi:10.1093/nar/25.17.3389 · Cited by 73,302
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204
Fast and sensitive protein alignment using DIAMOND
Buchfink, B., Xie, C., & Huson, D. H. (2015) · Nature Methods · doi:10.1038/nmeth.3176 · Cited by 14,201
Multi-scale context aggregation by dilated convolutions
Yu, F., & Koltun, V. (2015) · arXiv · doi:10.48550/arxiv.1511.07122 · Cited by 7,738
UniProt: the Universal Protein knowledgebase
Apweiler, R. (2003) · Nucleic Acids Research · doi:10.1093/nar/gkh131 · Cited by 7,593
Research works printed but not linked: 1
- El-Gebali, S., Richardson, L. & Finn, R. Repeats in Pfam. EMBL-EBI Training https://doi.org/10.6019/TOL.Pfam_repeats-t.2018.00001.1 (2018). · DOI
Cited by
Sorted by citations
Items 301–335 of 335
Graphical models for identifying pore‐forming proteins
Xu, N., Kahn, T. W., Jacob, T., & Liu, Y. (2024) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26687
Alignment-free sequence composition based features can distinguish paralogous proteinsPreprint
Machingal, P., Busi, R., Hemachandra, N., & Balaji, P. V. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.04.26.591291
A machine learning ride in the physics theme park: from quantum to biophysics
Requena, B. (2024) · UPCommons institutional repository (Universitat Politècnica de Catalunya) · doi:10.5821/dissertation-2117-409132
Protein Structure Classification in the Era of AlphaFold2
Waman, V., Dawson, N. L., Das, S., Lees, J., & Orengo, C. A. (2024) · Elsevier eBooks · doi:10.1016/b978-0-323-95502-7.00047-6
Chen, H., Zhang, B., Liu, K., Zheng, Z., Zhu, J., Li, Z., Liu, Y., Mu, J., & Wei, T. (2024) · Research Square · doi:10.21203/rs.3.rs-5450034/v1
Zulfiqar, R., Khan, T. A., Khan, A. A., Akhtar, R., & Ullah, S. (2025) · Discover Applied Sciences · doi:10.1007/s42452-025-07271-z · Cited by 1
“One of the most popular ensembles learning methods, Random Forest, iteratively builds a decision tree several times in an attempt to improve accuracy [10].”
Harnessing deep learning for proteome-scale detection of amyloid signaling motifs
Pysz, K., Gałązka, J., & Dyrka, W. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf200
ET-Pfam: Ensemble transfer learning for protein family predictionPreprint
Escudero, S., Duarte, S., Vitale, R., Fenoy, E., Bugnon, L. A., Milone, D. H., & Stegmayer, G. S. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.08.02.668111
“Pfam data was used for the experiments as in [2], where full sequences from 17,929 families of Pfam v.32.0 were split into train, development (dev) and test sets by clustering based on sequence identity.”
InfluentialDenger, A., & Helms, V. (2025) · Molecules · doi:10.3390/molecules30153226 · Cited by 1
“…as well as various predictive tasks, e.g., for secondary and tertiary structure [28–30], structural similarity and protein family membership [31,32], GO term annotations [33–37], protein–protein interactions (PPIs) [38,39], sequence variant effects [40,41], sub-cellular location [42,43], and…”
Li, Y., Wang, Z., Xing, R., Wang, X., & Gu, Y. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.132848
DomDiff: protein family and domain annotation via diffusion model and ESM2 embeddingPreprint
Zhang, C., Xia, H., & Peng, Y. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.10.28.685005
“…this task have focused on amino acid sequence similarity, roughly falling into three categories: (i) methods based on multiple sequence alignment (MSA)[9][10] (ii) methods based on genealogical clustering[11][12][13][14] and (iii) methods based on a deep learning model[15][16][17][18][19][20].”
Wynne, J. H., Azbijari, N., Thurber, A. R., & David, M. M. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2025.12.17.695047
“Recent studies have successfully used this approach and yielded functional discoveries by leveraging protein language models in antibiotic resistance [23], prokaryotic viral protein function [24,25], and phosphorous cycling protein families [26].”
Additional file 1 of Organizing the bacterial annotation space with amino acid sequence embeddings
Grigson, S. R., McKerral, J. C., Mitchell, J. G., & Edwards, R. A. (2022) · Figshare · doi:10.6084/m9.figshare.21200740
Additional file 2 of Protein language models can capture protein quaternary state
Avraham, O., Tsaban, T., Ben-Aharon, Z., Tsaban, L., & Schueler‐Furman, O. (2024) · Figshare · doi:10.6084/m9.figshare.24564257
Additional file 3 of Protein language models can capture protein quaternary state
Avraham, O., Tsaban, T., Ben-Aharon, Z., Tsaban, L., & Schueler‐Furman, O. (2024) · Figshare · doi:10.6084/m9.figshare.26639947
Artificial Intelligence Approaches for Constructing Digital Plant Proteomics
Dubey, P. K. A., Vijayarani, K., Manoj, M., Gavaskar, S., Robert, A. A., Prabhu, J. P. A., Anand, A. V., & Sumithra, A. (2026) · CABI biotechnology series. · doi:10.1079/9781800628649.0003
Fine-Tuning ESM2 for Predicting Ageing-Related Human Proteins
Xiao, G., Fan, C., Hu, Y., & Tseng, S.-P. (2026) · Smart innovation, systems and technologies · doi:10.1007/978-3-032-13048-8_26
Fenoy, E., Bugnon, L. A., Vitale, R., Duarte, S., Milone, D. H., & Stegmayer, G. S. (2026) · Journal of Structural Biology · doi:10.1016/j.jsb.2026.108294
Chuanyang, L., Liu, J., Qiu, X., Wu, X., Li, W., Min, L., Zhang, G., Zhang, S., & Zhu, L. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.20.706798 · Cited by 1
Evolutionary profile enhancement improves protein function annotation for remote homologsPreprint
Dai, S., Luo, J., & Luo, Y. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.03.03.709280
“Most existing methods are instantiated with deep neural networks 10,16,17,27–29 that can be decomposed into a feature extractor h ϕ and a linear classifier g ω , such that f θ = g ω ◦ h ϕ .”
Structural prediction of protein assemblies with deep learning
Bret, H. (2023) · theses.fr (ABES) · doi:10.70675/b3849091zce01z4ab0z91efz2b71d9cc754d
Rahimzadeh, F., Khanli, L. M., Salehpour, P., & Golabi, F. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.132559 · Cited by 1
Diagnosing protein sequence search in the era of language modelsPreprint
Zhou, H., Yang, Y., & Lu, Y. Y. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.04.26.720921
“As a result, PLM-based protein sequence search has been rapidly adopted, with several recent studies presenting PLM similarity as a powerful alternative to traditional alignment for large-scale search and remote homology detection [13–16].”
RNA Large Language Models in Virology
Shen, W., Zhu, L., Li, Y., & Huang, T. (2026) · Methods in molecular biology · doi:10.1007/978-1-0716-5264-0_17
Data-Efficient Exploration of Enzyme Function Using Family-Specific Machine LearningPreprint
Ahmed, F. H., Bender, A., Wijesinghe, A., Zhu, A. Y., Zhang, L., Gebbie, L. K., Marsh, A., Ishitate, C., Holdsworth, W., Jones, C., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.06.02.729712
“…sequence models demonstrate that neural networks trained on expansive sequence datasets can extract statistical representations directly from amino-acid sequence and support tasks such as gene annotation 9 , enzyme property prediction and functional inference 10–12 , and structural prediction 13 .”
Wang, Y., Zhao, Y., Wu, W., & Zhang, H. (2026) · Research Square · doi:10.21203/rs.3.rs-10102565/v1
Gu, J., Xu, Y., Sun, X., & Nie, Y. (2026) · ACS Synthetic Biology · doi:10.1021/acssynbio.6c00336 · Cited by 1
Tokenization in Protein Language Models: Methods, Taxonomy, and Applications
Linoubli, A., Abidi, K., & Farah, M. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-2597-2_11
Deployable high-fidelity metagenome binning at scale with QuickBin
Bushnell, B., & Villada, J. C. (2026) · Communications Biology · doi:10.1038/s42003-026-10782-z · Cited by 1
“Contaminated, fragmented, or marker-gene biased construction of MAGs risk introducing noise that can propagate hallucinations in generative models, or obscure subtle evolutionary signals 28,29,32 .”
Li, H., Zhou, D., Pu, W., Wu, G., Chun, X., Gao, X., & Yu, J. (2026) · Insects · doi:10.3390/insects17090928
“Deep learning-based protein-family classifiers already complement alignment-based annotation in broad settings [84].”
Hu, X.-Y., Ma, Y.-H., Ming, R., Zhang, H.-P., & Jiang, H.-J. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag476
“…biological technologies has driven the widespread application of feature (variable) selection across various fields, including target identification [ 1, 2], protein annotation [ 3], dimensionality reduction [ 4, 5], clinical phenotype prediction [ 6, 7], and disease-state characterization [ 8, 9].”
Shao, J., Wang, Z., Chen, Z., & Liu, B. (2026) · Science China Information Sciences · doi:10.1007/s11432-025-5060-9
“As proteins participate in nearly all life activities, such as metabolism, signal transduction, and molecular synthesis, accurate functional annotation is essential for studies in drug discovery and disease treatment [3].”
FAIRsharing record for: Integrated resource of protein families, domains and functional sites
FAIRsharing Team (2015) · FAIRsharing.org · doi:10.25504/fairsharing.pda11d
Abid, N., Qazi, O., Ahmed, H., Rafaqat, S., & Yousaf, M. (2026) · Diagnostics · doi:10.3390/diagnostics16213517
Metadata
- Authors
- Bileschi, M. L., Belanger, D., Bryant, D. H., Sanderson, T., Carter, B., Sculley, D., Bateman, A., DePristo, M. A., & Colwell, L. J.
- Affiliations
- Google (United States) (US)
- MIT Computer Science and Artificial Intelligence Laboratory (US)
- European Bioinformatics Institute (GB)
- Corresponding author
- Bileschi, Maxwell L.
- Colwell, Lucy J.
- ORCID
- Venue
- Nature Biotechnology
- Published
- 2022-02-21
- Volume
- 40
- Issue
- 6
- Pages
- 932-937
- ISSN
- 0733-222X (ISSN-L)
- 1087-0156
- 1546-1696
- OpenAlex
- W4213112325
- Cited by
- 332
- References
- 59
- Citations by year
- 2026: 39 · 2025: 66 · 2024: 70 · 2023: 73 · 2022: 45
- FWCI
- 20.90
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Locations
- Nature Biotechnology (publishedVersion)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion)
- PubMed (publishedVersion)
- Funders
- Wellcome Trust (210918/Z/18/Z)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Keywords
- protein function prediction, deep learning, Pfam, sequence annotation, remote homology detection, sequence clustering, microbial protein, human proteome, protein families
- MeSH terms
- Deep Learning, Amino Acid Sequence, Humans, Proteome, Databases, Protein, Proteomics, Molecular Sequence Annotation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
