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 101–200 of 335
Elucidating the functional roles of prokaryotic proteins using big data and artificial intelligence
Ardern, Z., Chakraborty, S., Lenk, F., & Kaster, A. (2023) · FEMS Microbiology Reviews · doi:10.1093/femsre/fuad003 · Cited by 22
“…Pfams CS EP/NEP FD SB GA GO SB EC EC, GO Pfams, Sbfam Giri et al. (2021) Kulmanov and Hoehndorf (2020) Du et al. (2020) Sommer and Salzberg (2021) Bileschi et al. (2019) Sandaruwan and Wannige (2021) Durrant et al. (2020) Zeng et al. (2019) Gainza et al. (2020) Zha et al. (2020) Zha et al.…”
BackgroundConstructing benchmark test sets for biological sequence analysis using independent set algorithms
Petti, S., & Eddy, S. R. (2022) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1009492 · Cited by 19
Wilburn, D. B., Shannon, A. E., Spicer, V., Richards, A., Yeung, D., Swaney, D. L., Krokhin, O. V., & Searle, B. C. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.05.30.542978 · Cited by 19
“Inspired by recent innovations in protein secondary structure prediction 24 and functional classification, 25 Chronologer learns to encode each amino acid type as an array of 64 numbers that, when tiled along a sequence dimension, creates a 2-dimensional matrix representation for each peptide.”
MethodJohnson, S. R., Peshwa, M., & Sun, Z. (2024) · eLife · doi:10.7554/elife.91415 · Cited by 17
“…annotation and remote homology detection rely on deep neural networks taking protein sequences as inputs and producing either a classification from a controlled vocabulary ( Bileschi et al., 2022 ; Sanderson et al., 2023 ), a natural language description ( Gane et al., 2022 ), positional…”
BackgroundMethodInfluentialThe Role of AI-Driven De Novo Protein Design in the Exploration of the Protein Functional Universe
Zhang, G., Liu, C., Lu, J., Zhang, S., & Zhu, L. (2025) · Biology · doi:10.3390/biology14091268 · Cited by 22
Sokhansanj, B. A., & ROSEN, G. L. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.105969 · Cited by 19
Magnúsdóttir, S., Saraiva, J., Bartholomäus, A., Soheili, M., Toscan, R. B., Zhang, J., & da Rocha, U. N. (2023) · Frontiers in Microbiology · doi:10.3389/fmicb.2023.1037845 · Cited by 18
“Current protein alignment methods fail to identify over a third of microbial proteins (Bileschi et al., 2022).”
BackgroundMethodFunctional annotation of proteins for signaling network inference in non-model species
den Broeck, L. V., Bhosale, D. K., Song, K., de Lima, C. F. F., Ashley, M. C. B., Zhu, T., Zhu, S., van de Cotte, B., Neyt, P., Ortiz, A. C., et al. (2023) · Nature Communications · doi:10.1038/s41467-023-40365-z · Cited by 18
“Speci fi cally, ProtCNN, a neural network trained on protein domain sequences, accurately annotated protein domains and improved and expanded on current proteins annotations 16 .”
InfluentialFunctions of intrinsically disordered proteins through evolutionary lenses
Pajkos, M., & Dosztányi, Z. (2021) · Progress in molecular biology and translational science · doi:10.1016/bs.pmbts.2021.06.017 · Cited by 19
Deep protein representations enable recombinant protein expression prediction
Martiny, H.-M., Armenteros, J. J. A., Johansen, A. R., Salomon, J., & Nielsen, H. (2021) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2021.107596 · Cited by 16
Schell, T., Greve, C., & Podsiadłowski, L. (2025) · Frontiers in Zoology · doi:10.1186/s12983-025-00561-7 · Cited by 24
ProtNote: a multimodal method for protein–function annotation
Char, S., Corley, N., Alamdari, S., Yang, K., & Amini, A. P. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf170 · Cited by 19
“…such as CLEAN and HiFi-NN (Ayres et al. 2025, Yu et al. 2023); ProtENN (Bileschi et al. 2022), which infers protein families from the Pfam database (Mistry et al. 2021); and ProteinBERT (Brandes et al. 2022), a GO-annotation-aware protein language model that can be fine-tuned on downstream tasks.”
High Bone Mass Disorders: New Insights From Connecting the Clinic and the Bench
Bergen, D. J. M., Maurizi, A., Formosa, M. M., McDonald, G. L. K., El‐Gazzar, A., Hassan, N., Brandi, M. L., Riancho, J. A., Rivadeneira, F., Ntzani, E., et al. (2020) · Journal of Bone and Mineral Research · doi:10.1002/jbmr.4715 · Cited by 17
“uk/interpro/).((167)) Advances in the coverage of Pfam also suggest that deep learning models will be a core component of future protein annotation tools and VUS interpretation.”
BackgroundKress, A., Poch, O., Lecompte, O., & Thompson, J. D. (2023) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2023.1178926 · Cited by 21
“This domain is also identi fi ed by a new Deep Learning method (Bileschi et al., 2022) that was trained on Pfam data and annotated as a Pfam-N (N for Network) domain in Interpro.”
MethodChang, D. H., Richardson, J. D., Lee, M., Lynn, D. M., Palecek, S. P., & Lehn, R. C. V. (2025) · Chemical Science · doi:10.1039/d4sc06689h · Cited by 16
Accelerating protein engineering with fitness landscape modelling and reinforcement learning
Sun, H., He, L., Deng, P., Liu, G., Zhao, Z., Jiang, Y., Cao, C., Ju, F., Wu, L., Liu, H., et al. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01103-w · Cited by 19
“The motif-level scoring module aims to extract consequential sequence patterns that are important for protein functions [28, 29].”
ProtSpace: A Tool for Visualizing Protein Space
Senoner, T., Olenyi, T., Heinzinger, M., Spannagl, A., Bouras, G. S., Rost, B., & Koludarov, I. (2025) · Journal of Molecular Biology · doi:10.1016/j.jmb.2025.168940 · Cited by 17
Twenty years of advances in prediction of nucleic acid-binding residues in protein sequences
Basu, S., Jing, Y., Kihara, D., & Kurgan, L. (2024) · Briefings Bioinform. · doi:10.1093/bib/bbaf016 · Cited by 15
Phytopathogen Effector Biology in the Burgeoning AI Era
Jones, D., & Raffaele, S. (2025) · Annual Review of Phytopathology · doi:10.1146/annurev-phyto-121823-081033 · Cited by 17
Protein–protein interaction and site prediction using transfer learning
Liu, T., Gao, H., Ren, X., Xu, G., Liu, B., Wu, N., Luo, H., Wang, Y., Tu, T., Yao, B., et al. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad376 · Cited by 15
PhaGenus: genus-level classification of bacteriophages using a Transformer model
Guan, J., Peng, C., Shang, J., Tang, X., & Sun, Y. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad408 · Cited by 15
Chow, C. F. W., Ghosh, S., Hadarovich, A., & Tóth-Petróczy, Á. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2401622121 · Cited by 15
Unifying the known and unknown microbial coding sequence spacePreprint
Vanni, C., Schechter, M. S., Acinas, S. G., Barberán, A., Buttigieg, P. L., Casamayor, E. O., Delmont, T. O., Duarte, C. M., Eren, A. M., Finn, R. D., et al. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.06.30.180448 · Cited by 15
AI-powered de novo architecturing of marine-mimetic antioxidant peptides in yeast synbio factories
Gao, L., Zhou, T., Yuan, J. X., Zhang, S., Hong, K., Zhang, Z., Hau, L. C., Cheang, W. S., Zhou, C.-C., & Wu, X. (2025) · Chemical Engineering Journal · doi:10.1016/j.cej.2025.169604 · Cited by 15
Evolutionary Context-Integrated Deep Sequence Modeling for Protein Engineering
Luo, Y., Vo, L. T., Ding, H., Su, Y., Liu, Y., Qian, W. W., Zhao, H., & Peng, J. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-45257-5_30 · Cited by 15
Frei, L., Gao, B., Han, J., Taft, J. M., Irvine, E. B., Weber, C. R., Kumar, R. K., Eisinger, B., Ignatov, A., Yang, Z., et al. (2025) · Nature Biomedical Engineering · doi:10.1038/s41551-025-01353-4 · Cited by 16
Fungtion: A Server for Predicting and Visualizing Fungal Effector Proteins
Li, J., Ren, J., Dai, W., Stubenrauch, C. J., Finn, R. D., & Wang, J. (2024) · Journal of Molecular Biology · doi:10.1016/j.jmb.2024.168613 · Cited by 13
Comparative Assessment of Protein Large Language Models for Enzyme Commission Number Prediction
Capela, J., Zimmermann‐Kogadeeva, M., van Dijk, A. D. J., de Ridder, D., Dias, Ó., & Rocha, M. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06081-9 · Cited by 13
“For example, current alignment-based tools are unable to determine the functions of one-third of microbial protein sequences [35].”
How Deep Learning Tools Can Help Protein Engineers Find Good Sequences
Osadchy, M., & Kolodny, R. (2021) · The Journal of Physical Chemistry B · doi:10.1021/acs.jpcb.1c02449 · Cited by 16
Fujita, S., & Terada, T. (2024) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2024.11.028 · Cited by 13
“Various models have been developed to predict GO terms [9], EC classes [10–12], and functional domains within protein sequences [13].”
ProtGO: universal protein function prediction utilizing multi-modal gene ontology knowledge
Wang, B., Geng, Y.-A., Cheng, X., Chen, B., Bei, Z., Wang, W., Tang, J., & Song, L. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf390 · Cited by 15
Transformer Neural Networks for Protein Family and Interaction Prediction Tasks
Nambiar, A., Liu, S., Heflin, M., Forsyth, J. M., Maslov, S., Hopkins, M., & Ritz, A. (2022) · Journal of Computational Biology · doi:10.1089/cmb.2022.0132 · Cited by 68
“Guided by a deep learning framework, recent models based on convolutional neural network (CNN) and recurrent neural network (RNN) architectures have been successful in achieving state-of-the-art accuracy on the protein family classification task (Heinzinger et al., 2019; Bileschi et al., 2022).”
ApHMM: Accelerating Profile Hidden Markov Models for Fast and Energy-efficient Genome Analysis
Fırtına, C., Pillai, K., Kalsi, G. S., Suresh, B., Cali, D. S., Kim, J. S., Shahroodi, T., Cavlak, M. B., Lindegger, J., Alser, M., et al. (2023) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3632950 · Cited by 12
“Classifying protein sequences into families is widely used to analyze the potential functions of the proteins of interest [15, 47, 81, 106, 119, 121].”
BackgroundDecoding the Molecular Language of Proteins with EvollaPreprint
Zhou, X., Han, C., Zhang, Y., Du, H., Tian, J., Su, J., Liu, R., Zhuang, K., Jiang, S., Gitter, A., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.01.05.630192 · Cited by 16
Mu, Q., Yu, G., Zhou, G., He, Y., & Zhang, J. (2025) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqaf058 · Cited by 11
“For example, DeepFam [ 36 ] and ProtCNN [ 37 ] utilized neural networks to process protein sequences for non-hierarchical classification of protein families.”
Krzynowek, A., Snoeks, J., & Faust, K. (2026) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013892 · Cited by 12
Enzymes for consumer products to achieve climate neutrality
Molina‐Espeja, P., Sanz‐Aparicio, J., Golyshin, P. N., Robles‐Martín, A., Guallar, V., Beltrametti, F., Müller, M., Yakimov, M. M., Modregger, J., van Logchem, M., et al. (2023) · Oxford Open Climate Change · doi:10.1093/oxfclm/kgad003 · Cited by 12
Biophysics-guided uncertainty-aware deep learning uncovers high-affinity plastic-binding peptides
Alshehri, A. S., Bergman, M. T., You, F., & Hall, C. K. (2025) · Digital Discovery · doi:10.1039/d4dd00219a · Cited by 12
Deep learning methods for protein representation and function prediction: A comprehensive overview
Wang, M., Nie, Z., He, Y., Vasilakos, A. V., Cheng, Q. S., & Ren, Z. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110977 · Cited by 11
UDSMProt: Universal Deep Sequence Models for Protein ClassificationPreprint
Strodthoff, N., Wagner, P., Wenzel, M., & Samek, W. (2019) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/704874 · Cited by 14
Kök, B. Ö., Altunoğlu, Y. Ç., Öncül, A. B., Karacı, A., & Baloğlu, M. C. (2023) · Journal of Bioinformatics and Computational Biology · doi:10.1142/s0219720023500154 · Cited by 10
Zhang, G., Wang, H., Zhang, Z., Zhang, L., Guo, G., Yang, J., Yuan, F., & Ju, F. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae319 · Cited by 12
“These models, known as protein language models (PLMs), excel in learning comprehensive and sophisticated semantic representations that establish meaningful connections between gene sequences and protein function [25, 26].”
BackgroundIn silico prediction of variant effects: promises and limitations for precision plant breeding
Sendrowski, J., Bataillon, T. M., & Ramstein, G. P. (2025) · Theoretical and Applied Genetics · doi:10.1007/s00122-025-04973-1 · Cited by 12
“To alleviate this bottleneck, computational approaches may provide additional information about the existence and function of gene isoforms (Bileschi et al. 2022; Schulz et al. 2023).”
John, C., Sahoo, J., Madhavan, M., & Mathew, O. K. (2023) · Current Bioinformatics · doi:10.2174/1574893618666230320103421 · Cited by 14
Intelligent Protein Design and Molecular Characterization Techniques: A Comprehensive Review
Wang, J., Chen, C., Yao, G., Ding, J., Wang, L., & Jiang, H. (2023) · Molecules · doi:10.3390/molecules28237865 · Cited by 10
“In February 2022, Bileschi et al. [44] used a dataset constructed from the Protein Families Database (Pfam) to train a neural network (called ProtCNN) to functionally classify protein sequences to achieve a 200-fold increase in speed, and a 9-fold reduction in error, compared to the traditional…”
MethodDohan, D., Gane, A., Bileschi, M. L., Belanger, D., & Colwell, L. J. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467163 · Cited by 13
“These deep learning approaches to protein sequence annotation use 1D convolutional neural networks with residual connections (ResNets) [7].”
BackgroundMethodInfluentialSelf-citationA hybrid deep learning model for classification of plant transcription factor proteins
Öncül, A. B., & Çelik, Y. (2022) · Signal Image and Video Processing · doi:10.1007/s11760-022-02419-5 · Cited by 10
“…et al.’s deep learning studies [12], UDSMProt [13], LSTM and GRU-based studies by Le et al. [14], LSTM-based studies by Li et al. [15], CNN and LSTM-based studies by Bileshi et al. [16], TAPE [17], Belzen et al. [18], Torissi et al. deep learning study [19] and the work of Gustafsson et al. [20].”
BackgroundMethodLearning sequence, structure, and function representations of proteins with language modelsPreprint
Hamamsy, T., Barot, M., Morton, J. T., Steinegger, M., Bonneau, R., & Cho, K. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.11.26.568742 · Cited by 12
“…function prediction and annotation that also include a wide range of data beyond sequence [11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25], including methods that use protein sequences [15, 18, 19, 21], protein-protein interaction networks [15, 26, 27], protein structures [17], and…”
MethodMulti-Modal Large Language Model Enables Protein Function PredictionPreprint
Huo, M., Guo, H. W., Cheng, X., Singh, D., Rahmani, H., Li, S., Gerlof, P., Ideker, T., Grotjahn, D. A., Villa, E., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.08.19.608729 · Cited by 12
“Predicting protein functions is a complex and challenging task due to the inherent diversity and intricate nature of proteins (6–10).”
Johnson, S. R., Weigele, P. R., Fomenkov, A., Ge, A., Vincze, A., Eaglesham, J. B., Roberts, R. J., & Sun, Z. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkae1175 · Cited by 10
Unsupervised Deep Learning Can Identify Protein Functional Groups from Unaligned Sequences
David, K. T., & Halanych, K. M. (2023) · Genome Biology and Evolution · doi:10.1093/gbe/evad084 · Cited by 9
Chen, H., Lu, D. X., Xiao, Z., Li, S., Zhang, W., Luan, X., Zhang, W., & Zheng, G. (2024) · Health Information Science and Systems · doi:10.1007/s13755-024-00300-y · Cited by 10
Cutting-edge deep-learning based tools for metagenomic research
Karin, E. L., & Steinegger, M. (2025) · National Science Review · doi:10.1093/nsr/nwaf056 · Cited by 10
Schmitz, M. A., Dimonaco, N. J., Clavel, T., & Hitch, T. C. A. (2025) · Nature Communications · doi:10.1038/s41467-025-58442-w · Cited by 10
“MiProGut provides an improved resource for protein sequence identi fi cation, but functional annotation is required to understand the role of these proteins, both in relation to the microbiome, and their impact on the host 40,41 .”
Protein sequence classification using natural language processing techniques
Perveen, H., & Weeds, J. (2025) · Discover Artificial Intelligence · doi:10.1007/s44163-025-00304-x · Cited by 9
Accelerating Biological Insight for Understudied Genes
Reynolds, K. A., Rosa‐Molinar, E., Ward, R. E., Zhang, H., Urbanowicz, B. R., & Settles, A. M. (2021) · Integrative and Comparative Biology · doi:10.1093/icb/icab029 · Cited by 8
Transfer learning improves antibiotic resistance class predictionPreprint
Hamid, M.-N., & Friedberg, I. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.04.17.047316 · Cited by 9
“Transfer learning has been shown to be effective in various tasks of protein sequence classification [20, 21, 22, 23] where there is a small number of labeled data.”
BackgroundSEQUENCE VS. STRUCTURE: DELVING DEEP INTO DATA-DRIVEN PROTEIN FUNCTION PREDICTIONPreprint
Tian, X., Wang, Z., Yang, K., Su, J., Du, H., Zheng, Q., Guo, G., Yang, M., Yang, F., & Yuan, F. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.04.02.534383 · Cited by 8
“…HTS has produced a deluge of unmeasured protein data in diverse species and spawned a series of faster and cheaper computational prediction methods (Bileschi et al., 2022; Radivojac et al., 2013; Rives et al., 2021b; Dallago et al., 2021; Pazos & Sternberg, 2004; Hie et al., 2021; Unsal et al.,…”
Castaño, J. D., Khoury, I. E., Goering, J., Evans, J. E., & Zhang, J. (2024) · Applied and Environmental Microbiology · doi:10.1128/aem.00122-24 · Cited by 8
Artificial intelligence in protein-based detection and inhibition of AMR pathways
Sadhukhan, S., Bhattacharya, R., Bhattcharya, D., Sahana, S., Pradhan, B., Pandit, S., Gill, H. S., Rajeev, M., Nag, M., & Lahiri, D. (2025) · Journal of Computer-Aided Molecular Design · doi:10.1007/s10822-025-00710-4 · Cited by 10
Kennedy, E. N., Foster, C. A., Barr, S. A., & Bourret, R. B. (2022) · Biochemical Society Transactions · doi:10.1042/bst20220849 · Cited by 8
ProteinEngine: Empower LLM with Domain Knowledge for Protein Engineering
Shen, Y., Lv, O., Zhu, H., & Wang, Y. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-66538-7_37 · Cited by 8
The substrate scopes of enzymes: a general prediction model based on machine and deep learningPreprint
Kroll, A., Ranjan, S., Engqvist, M. K. M., & Lercher, M. J. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.05.24.493213 · Cited by 8
“Deep learning models have been used to predict enzyme functions by either predicting their assignment to EC classes 19,20,21 , or by predicting functional domains within the protein sequence 22 .”
MethodProtein language models uncover carbohydrate-active enzyme function in metagenomics
Thurimella, K., Mohamed, A. M. T., Li, C., Vatanen, T., Graham, D. B., Owens, R. M., La Rosa, S. L., Plichta, D. R., Bacallado, S., & Xavier, R. J. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06286-y · Cited by 9
“The use of pLMs has emerged as a powerful tool for unraveling protein functions in microbial genomics [27, 28, 34], and our results further emphasize their efficacy in this context.”
DeepMethyGene: a deep-learning model to predict gene expression using DNA methylations
Yan, Y., Chai, X., Liu, J., Wang, S., Li, W., & Huang, T. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06115-2 · Cited by 9
Large-scale protein clustering in the age of deep learning
Pereira, J., Pantolini, L., Durairaj, J., & Schwede, T. F. (2025) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2025.103078 · Cited by 7
Approaches for accelerating microbial gene function discovery using artificial intelligence
Palsson, B. Ø., Lee, S. Y., & Kim, G. B. (2026) · Nature Microbiology · doi:10.1038/s41564-025-02214-1 · Cited by 8
“…which refers to identifying evolutionary relationships between proteins with very low sequence similarity, can also be facilitated by combining a deep learning model that predicts the Pfam family—a widely used classification system for protein domains and families—with homology analysis 32 .”
Predrag Radivojac, A. (2022) · Cell Systems · doi:10.1016/j.cels.2022.05.006 · Cited by 7
Feng, J., Wang, Z., Li, Y., Ding, B., Wei, Z., & Xu, H. (2022) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3511808.3557395 · Cited by 8
“Inspired by these successes, researchers have recently attempted to apply deep learning to solving scientific problems, namely, AI for science [3, 23], where molecular property prediction is among the most critical and promising ones.”
BackgroundMapping the FtsQBL divisome components in bacterial NTD pathogens as potential drug targets
Kaur, H., & Lynn, A. M. (2023) · Frontiers in Genetics · doi:10.3389/fgene.2022.1010870 · Cited by 8
“(Bileschi et al., 2022), and since been applied to the uniprotKB database with higher accuracy, functionally identifying ctFtsQ and ctFtsL, but is still unable to annotate the FtsB and FtsL homologs fromMycobacterium and ctFtsB.”
MethodClassification of G protein-coupled receptors using attention mechanism
Paki, R., Nouranı, E., & Farajzadeh, D. (2020) · Gene Reports · doi:10.1016/j.genrep.2020.100882 · Cited by 7
Frei, L., Gao, B., Han, J., Taft, J. M., Irvine, E. B., Weber, C. R., Kumar, R. K., Eisinger, B. N., & Reddy, S. T. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.10.09.561492 · Cited by 7
Protein Language Models Uncover Carbohydrate-Active Enzyme Function in Metagenomics
Thurimella, K., Mohamed, A. M. T., Graham, D. B., Owens, R. M., La Rosa, S. L., Plichta, D. R., Bacallado, S., & Xavier, R. J. (2023) · doi:10.1101/2023.10.23.563620 · Cited by 7
“The use of pLMs has emerged as a powerful tool for unraveling protein functions in microbial genomics 28–30 , and our results further emphasize their efficacy in this context.”
BackgroundMethodProtEx: A Retrieval-Augmented Approach for Protein Function PredictionPreprint
Shaw, P., Gurram, B., Belanger, D., Gane, A., Bileschi, M. L., Colwell, L. J., Toutanova, K., & Parikh, A. P. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.05.30.596539 · Cited by 8
“We compare ProtEx with two strong homology-based approaches (BLAST and Top pick HMM, as described in Dohan et al. [16]), ProtENN [9] a convolutional neural network ensemble, and ProtNLM [21] and ProtTNN [16], which are pretrained Transformer models.”
BackgroundMethodSelf-citationProteomic advance of ischemic stroke: preclinical, clinical, and intervention
Zhao, T., Zeng, J., Zhang, R., Pu, L., Wang, H., Pan, L., Jiang, Y., Dai, X., Sha, Y., & Han, L. (2023) · Metabolic Brain Disease · doi:10.1007/s11011-023-01262-y · Cited by 6
Albin, D., Ramsahoye, M., Kochavi, E., & Alistar, M. (2024) · Frontiers in Microbiology · doi:10.3389/fmicb.2024.1446097 · Cited by 6
Chisompola, D., Luwaya, E. L., Nzobokela, J., Mwansa, P., & Chakulya, M. (2026) · Frontiers in Microbiology · doi:10.3389/fmicb.2025.1717859 · Cited by 10
“Convolutional and pooling layers then generate feature maps that summarize motif occurrence and position, conferring robustness to sequence variability (Unsal et al., 2022; Bileschi et al., 2022).”
InfluentialGolenko, Y., Ismailova, A., Shaushenova, A., Mutalova, Z., Dossalyanov, D., Ainagulova, A., & Naizagarayeva, A. (2022) · Eastern-European Journal of Enterprise Technologies · doi:10.15587/1729-4061.2022.263270 · Cited by 6
Tamir, A., Salem, M., & Yuan, J. S. (2023) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2023.3311427 · Cited by 6
Transfer learning: The key to functionally annotate the protein universe
Bugnon, L. A., Fenoy, E., Edera, A., Raad, J., Stegmayer, G. S., & Milone, D. H. (2023) · Patterns · doi:10.1016/j.patter.2023.100691 · Cited by 5
“Indeed, it has been an important step toward overcoming this limitation, 4 and we show that this issue can be further significantly reduced with transfer learning (TL) by transferring representations of protein sequences already learned without requiringannotationsfromlarge-scalepro-tein data.”
BackgroundLSTM-GRU Based Deep Learning Model with Word2Vec for Transcription Factors in Primates
Öncül, A. B. (2023) · Balkan Journal of Electrical and Computer Engineering · doi:10.17694/bajece.1191009 · Cited by 6
Perspectives in computational mass spectrometry: recent developments and key challenges
Sachsenberg, T., Pino, L. K., Brunet, M., Bludau, I., Kohlbacher, O., Vizcaino, J. A., & Bittremieux, W. (2024) · Bioinformatics Advances · doi:10.1093/bioadv/vbaf301 · Cited by 6
Fusion/fission protein family identification in Archaea
Padalko, A., Nair, G., & Sousa, F. L. (2024) · mSystems · doi:10.1128/msystems.00948-23 · Cited by 6
Imam, I. A., Bailey, S., Wang, D., Zeng, S., Xu, D. H., & Shao, Q. (2024) · Langmuir · doi:10.1021/acs.langmuir.4c04140 · Cited by 5
Machine and deep learning to predict viral fusion peptides
Sequeira, A. M., Rocha, M., & Lousa, D. (2025) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2025.02.011 · Cited by 5
“However, this method typically requires peptides or proteins to be of the same length for compatibility [11,15,16,20].”
Zhang, Y., Luo, J., Li, H., Li, W.-R., Li, W., Zhang, M., & Zhang, G. (2025) · Macromolecules · doi:10.1021/acs.macromol.5c02077 · Cited by 5
Compressing the collective knowledge of ESM into a single protein language model
Dinh, T., Jang, S.-K., Zaitlen, N., & Ntranos, V. (2026) · Nature Methods · doi:10.1038/s41592-026-03050-9 · Cited by 7
Elgad, N., Babayew, R., Last, M., Weiss, A. M., Gilad, E., Levy, G. K., & Halevy, I. (2024) · Journal of Radioanalytical and Nuclear Chemistry · doi:10.1007/s10967-024-09461-2 · Cited by 5
ThermoFinder: A sequence-based thermophilic proteins prediction framework
Yu, H., & Luo, X. (2024) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2024.132469 · Cited by 5
Luo, J., & Luo, Y. (2022) · Pacific Symposium on Biocomputing (PSB) · doi:10.1142/9789811270611_0011 · Cited by 5
耿, 浩. (2024) · Réngōng zhìnéng qiányán yǔ yìngyòng · doi:10.52810/faai.2024.005 · Cited by 5
Recurrent Deep Neural Networks for Enzyme Functional Annotation
Sequeira, A. M., & Rocha, M. (2021) · Lecture notes in networks and systems · doi:10.1007/978-3-030-86258-9_7 · Cited by 5
Poincaré maps for visualization of large protein families
Klimovskaia, A., Ren, Y., Meersche, Y. V., Gelly, J., & Galochkina, T. V. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad103 · Cited by 5
Tuned Fitness Landscapes for Benchmarking Model-Guided Protein DesignPreprint
Thomas, N., Agarwala, A., Belanger, D., Song, Y. S., & Colwell, L. J. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.10.28.514293 · Cited by 5
Self-citationWolf, B., Shehu, P., Brenker, L., Bachmann, A., Kroell, A.-S. H., Southern, N. T., Holderbach, S., Eigenmann, J., Aschenbrenner, S., Mathony, J., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.12.04.626757 · Cited by 5
Beyond sequence similarity: toward function-based screening of nucleic acid synthesis
Abel, G. R., Alexanian, T., Bartling, C., Beal, J., Curtis, S., Flyangolts, K., Foner, L. N., Forry, S. P., Godbold, G. D., Horvitz, E., et al. (2026) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2026.1832724 · Cited by 6
Optimizing Scorpion Toxin Processing through Artificial Intelligence
Psenicnik, A., Ojanguren-Affilastro, A. A., Graham, M. R., Hassan, M. K., Abdel-Rahman, M. A., Sharma, P. P., & Santibáñez‐López, C. E. (2024) · Toxins · doi:10.3390/toxins16100437 · Cited by 4
“Bileschi et al. [25] used deep learning models trained on sequences from the Pfam database to predict functional annotations for unknown sequences, extending the coverage of the Pfam database by >9.5%.”
MethodInfluentialTMVisDB: Annotation and 3D-visualization of Transmembrane Proteins
Olenyi, T., Marquet, C., Grekova, A., Houri, L., Heinzinger, M., Dallago, C., & Rost, B. (2025) · Journal of Molecular Biology · doi:10.1016/j.jmb.2025.168997 · Cited by 4
DeepSEA: an alignment-free explainable approach to annotate antimicrobial resistance proteins
Borelli, T. C., Paschoal, A. R., & Silva, R. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06256-4 · Cited by 4
“Deep learning-based models have emerged as a solution to this issue, as neural networks can learn complex and nonlinear rules from data and use them to annotate proteins without direct sequence alignment to known references (Bileschi et al., 2022).”
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
