Cited by 1,131Protein Structure and DynamicsMachine Learning in BioinformaticsRNA and protein synthesis mechanisms

The DOI matches its Crossref record; 80 of the 80 references it cites were found in a registry or an index.

Unified rational protein engineering with sequence-based deep representation learning

Open access

Alley, Ethan C., Khimulya, Grigory, Biswas, Surojit, AlQuraishi, Mohammed, Church, George M.·2019·Nature Methods

DOI 10.1038/s41592-019-0598-1PMC PMC7067682

Published version of a preprint

Open this paper in CoChat — free account Open PDF

Abstract

Rational protein engineering requires a holistic understanding of protein function. Here, we apply deep learning to unlabeled amino-acid sequences to distill the fundamental features of a protein into a statistical representation that is semantically rich and structurally, evolutionarily and biophysically grounded. We show that the simplest models built on top of this unified representation (UniRep) are broadly applicable and generalize to unseen regions of sequence space. Our data-driven approach predicts the stability of natural and de novo designed proteins, and the quantitative function of molecularly diverse mutants, competitively with the state-of-the-art methods. UniRep further enables two orders of magnitude efficiency improvement in a protein engineering task. UniRep is a versatile summary of fundamental protein features that can be applied across protein engineering informatics.

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

Crossref · 10.1038/s41592-019-0598-1

Cites

80 of 80 references verified0 citation errors · 0 not checkable

Statistics

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?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 1×
“We used basic agglomerative clustering (implemented in Python with sklearn 73 ) with the average linkage metric and Euclidean distance for all vector representations, and a precomputed Levenshtein distance matrix using the Python Levenshtein package for the sequence-only control.”
Method
The Protein Data Bank
Berman, H. M. (2000) · Nucleic Acids Research · doi:10.1093/nar/28.1.235 · Cited by 40,196 · cited 4×
“We computed the full sequence of hidden states at every position of 3,720 single-domain proteins from the SCOP database 60 (for which we had secondary structure annotation for each of the positions obtained from the PDB 80 ).”
MethodInfluential
Untitled work
(2008) · Journal of Machine Learning Research · Cited by 33,051
Dictionary of protein secondary structure: Pattern recognition of hydrogen‐bonded and geometrical features
Kabsch, W., & Sander, C. (1983) · Biopolymers · doi:10.1002/bip.360221211 · Cited by 15,320
On the Properties of Neural Machine Translation: Encoder–Decoder Approaches
Cho, K., van Merriënboer, B., Bahdanau, D., & Bengio, Y. (2014) · SSST@EMNLP · doi:10.3115/v1/w14-4012 · Cited by 7,251
Learning to Forget: Continual Prediction with LSTM
Gers, F. A., Schmidhuber, J., & Cummins, F. (2000) · Neural Computation · doi:10.1162/089976600300015015 · Cited by 6,122
NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations
Valiev, M., Bylaska, E. J., Govind, N., Kowalski, K., Straatsma, T. P., van Dam, H. J. J., Wang, D., Nieplocha, J., Aprà, E., Windus, T. L., et al. (2010) · Computer Physics Communications · doi:10.1016/j.cpc.2010.04.018 · Cited by 5,849
In-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537
An Overview of Multi-Task Learning in Deep Neural NetworksPreprint
Ruder, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.05098 · Cited by 3,827
Molecular dynamics simulations of biomolecules
Karplus, M., & McCammon, J. (2002) · Nature Structural Biology · doi:10.1038/nsb0902-646 · Cited by 3,401

Show all 80 references

Research works printed but not linked: 8

Brookes, D. H., Park, H. & Listgarten, J. Conditioning by adaptive sampling for robust design. Proc. Machine Learn. Res. 97, 773–782 (2019).
Hernández-Lobato, J. M., Requeima, J., Pyzer-Knapp, E. O. & Aspuru-Guzik, A. Parallel and distributed thompson sampling for large-scale accelerated exploration of chemical space. Preprint at arXiv https://arxiv.org/abs/1706.01825 (2017). · arXiv
Snoek, J., Larochelle, H. & Adams, R. P. in Advances in Neural Information Processing Systems Vol. 25 (eds. Pereira, F. et al.) 2951–2959 (Curran Associates, Inc., 2012).
Griffiths, R.-R. & Hernández-Lobato, J. M. Constrained Bayesian optimization for automaticchemical design. Preprint at arXiv https://arxiv.org/abs/1709.05501 (2017). · arXiv
Yang, K. K., Chen, Y., Lee, A. & Yue, Y. Batched stochastic Bayesian optimization via combinatorial constraints design. Preprint at arXiv https://arxiv.org/abs/1904.08102 (2019). · arXiv
González, J., Longworth, J., James, D. C. & Lawrence, N. D. Bayesian optimization for synthetic gene design. Preprint at arXiv https://arxiv.org/abs/1505.01627 (2015). · arXiv
Jouppi, N. P. et al. In-datacenter performance analysis of a tensorprocessing unit. In Proc. 44th Annual International Symposium of Computer Architecture Vol. 45, 1–12 (ACM, 2017).
Salimans, T. & Kingma, D. P. Weight normalization: a simple reparameterization to accelerate training of deep neural networks. Preprint at arXiv https://arxiv.org/abs/1602.07868 (2016). · arXiv

Sorted by citations

Items 801–900 of 1,003

Identification of plant vacuole proteins by using graph neural network and contact maps
Sui, J., Chen, J., Chen, Y., Iwamori, N., & Sun, J. (2023) · BMC Bioinformatics · doi:10.1186/s12859-023-05475-x · Cited by 3
“This model utilized the deep learning embedding methods UniRep [30] and SeqVec [31] to extract the properties of peroxisomal proteins.”
Method
Protein Sorting Prediction
Nielsen, H. (2023) · Methods in molecular biology · doi:10.1007/978-1-0716-3445-5_2 · Cited by 3
EGGNet, a Generalizable Geometric Deep Learning Framework for Protein Complex Pose Scoring
Wang, Z., Brand, R., Adolf-Bryfogle, J., Grewal, J., Qi, Y., Combs, S. A., Golovach, N., Alford, R., Rangwala, H., & Clark, P. M. (2024) · ACS Omega · doi:10.1021/acsomega.3c04889 · Cited by 3
Subtask-Aware Representation Learning for Predicting Antibiotic Resistance Gene Properties via Gating-Controlled Mechanism
Zhao, W., Wu, J., Luo, S., Jiang, X., He, T., & Hu, X. T. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3390246 · Cited by 3
“For comprehensively understanding ARGs, many machine learning-based methods [5], [7], [16], [19], [25] have been proposed for predicting properties of ARGs.”
AbGraftBERT: Enhancing Antibody Design Through Self-Grafting and Amino Acid-Based Perturbation of Pre- Trained Language Models
Wang, Z., Li, X.-G., Xie, Y., Wen, Z., Jin, R., Takada, H., & Nagatomi, R. (2024) · IEEE Access · doi:10.1109/access.2024.3416461 · Cited by 3
Evolutionary Probability and Stacked Regressions Enable Data-Driven Protein Engineering with Minimized Experimental Effort
Illig, A.-M., Siedhoff, N. E., Davari, M. D., & Schwaneberg, U. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c00704 · Cited by 3
PRONTO-TK: a user-friendly PROtein Neural neTwOrk tool-kit for accessible protein function prediction
Politano, G. M. M., Benso, A., Rehman, H. U., & Re, A. (2024) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqae112 · Cited by 3
An augmented transformer model trained on protein family specific variant data leads to improved prediction of variants of uncertain significance
Joshi, D., Pradhan, S., Sajeed, R., Srinivasan, R., & Rana, S. (2025) · Human Genetics · doi:10.1007/s00439-025-02727-z · Cited by 3
“Evolutionary fine-tuning or evotun - ing was first introduced in the UniRep paper (Alley et al. 2019 ), where the pretrained mLSTM weights get fine-tuned through weight updates using sequences homologous to the protein of interest.”
AI-driven parametrization of Michaelis–Menten maximal velocity: Advancing in silico new approach methodologies (NAMs)
Karakoltzidis, A., Karakitsios, S., & Sarigiannis, D. A. (2025) · NAM journal. · doi:10.1016/j.namjnl.2025.100012 · Cited by 3
“However, upon incorporating enzyme representations — utilizing the model by Alley et al. (2019) for enzyme representations — the performance improved, achieving an R-squared of 0.45 on unseen data points.”
Influential
Leveraging pretrained deep protein language model to predict peptide collision cross section
Nakai-Kasai, A., Ogata, K., Ishihama, Y., & Tanaka, T. (2024) · bioRxiv · doi:10.1038/s42004-025-01540-z · Cited by 4
“A number of deep protein language models for general purpose [25, 26, 27, 28, 29, 30, 31] and specific tasks [32, 33] have so far been proposed.”
Trainable subnetworks reveal insights into structure knowledge organization in protein language models
Vinod, R., Amini, A. P., Crawford, L. A., & Yang, K. (2026) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013925 · Cited by 2
“Many PLMs are pretrained with a self-supervised masked language modeling (MLM) ob-25 jective, where models are tasked with predicting the amino acid identities of randomly masked tokens in a sequence [1, 2, 14, 15].”
The Temperature Prediction of Hydro-generating Units based on Temporal Convolutional Network and Recurrent Neural Network
Li, J., Wu, Y., Li, Y., Ji, X., & Zheng, B. (2021) · Chinese Control Conference (CCC) · doi:10.23919/ccc52363.2021.9549853 · Cited by 3
“In recent years, temperature prediction has been greatly improved, deep learning network has been used in the time series prediction [9].”
Background
SIPF: Sampling Method for Inverse Protein Folding
Fu, T., & Sun, J. (2022) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3534678.3539284 · Cited by 3
“There is also a parallel research line on (unconditional) amino acid sequence learning, including protein sequence representation learning [2, 5, 6, 17, 18, 26] and protein sequence generation [1, 21, 24, 27, 34, 36, 39].”
BackgroundMethod
R-PCR: Recurrent Point Cloud Registration Using High-Order Markov Decision
Cheng, X., Yan, S., Liu, Y., Zhang, M., & Chen, C. (2023) · Remote Sensing · doi:10.3390/rs15071889 · Cited by 3
Attribution of genetic engineering: A practical and accurate machine-learning toolkit for biosecurityPreprint
Alley, E. C., Turpin, M., Liu, A. B., Kulp-McDowall, T., Swett, J. L., Edison, R., Stetina, S. E. V., Church, G. M., & Esvelt, K. M. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.08.22.262576 · Cited by 3
Self-citation
GLAMOUR: Graph learning over macromolecule representations
Mohapatra, S., An, J., & Gómez‐Bombarelli, R. (2021) · doi:10.1021/scimeetings.1c01022 · Cited by 3
“Machine learning (ML) applications to individual macromolecule classes, such as proteins, have been very successful but typically rely on sequence-based representations that effectively capture structures with linear architectures (Schissel et al., 2021; Alley et al., 2019).”
Background
LM-GVP: A Generalizable Deep Learning Framework for Protein Property Prediction from Sequence and StructurePreprint
Wang, Z., Combs, S. A., Brand, R. M., Calvo, M. R., Xu, P., Price, G., Golovach, N., Salawu, E. O., Wise, C. J., Ponnapalli, S. P., et al. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.09.21.460852 · Cited by 3
Proteochemometric Models Using Multiple Sequence Alignments and a Subword Segmented Masked Language ModelPreprint
Gaspar, H. A., Ahmed, M. A., Edlich, T., Fabian, B., Varszegi, Z., Segler, M., Meyers, J., & Fiscato, M. (2021) · ChemRxiv · doi:10.26434/chemrxiv.14604720.v1 · Cited by 3
CELL-E: A Text-To-Image Transformer for Protein Localization PredictionPreprint
Khwaja, E., Song, Y. S., & Huang, B. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.05.27.493774 · Cited by 4
“UniRep is an LSTM model that predicts the next amino acid in a variable length sequence [42], while BERT uses bidirectional masked language modeling to predict the identity of masked tokens throughout the sequence [43].”
Method
OpenProteinSet: Training data for structural biology at scale.Preprint
Ahdritz, G., Bouatta, N., Kadyan, S., Jarosch, L., Berenberg, D., Fisk, I., Watkins, A. M., Ra, S., Bonneau, R., & AlQuraishi, M. (2023) · PubMed · Cited by 3
Self-citation
Design of Enzymes for Biocatalysis, Bioremediation, and Biosensing using Variational Autoencoder-Generated Latent SpacesPreprint
Kohout, P., Vasina, M., Majerova, M., Novakova, V., Damborský, J., Bednář, D., Marek, M., Prokop, Z., & Mazurenko, S. (2023) · ChemRxiv · doi:10.26434/chemrxiv-2023-jcds7 · Cited by 3
Complex-based Ligand-Binding Proteins Redesign by Equivariant Diffusion-based Generative ModelsPreprint
Nguyen, V. T. D., Nguyen, N. D., & Hy, T. (2024) · bioRxiv · doi:10.1101/2024.04.17.589997 · Cited by 2
“Key traditional approaches include: • Rational Design 12 focused on introducing specific mutations into proteins based on known structural and functional insights.”
Background
Rational engineering of lipid-binding probes via high-throughput protein-lipid interaction screeningPreprint
Nishimura, T., Tsuboyama, K., Nakagaki, Y., Yamamoto, E., & Mizushima, N. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.01.05.627504 · Cited by 3
SENSE-PPI reconstructs interactomes within, across, and between species at the genome scale
Volzhenin, K., Bittner, L., & Carbone, A. (2024) · iScience · doi:10.1016/j.isci.2024.110371 · Cited by 4
Semi-supervised prediction of protein fitness for data-driven protein engineering
Olivares‐Gil, A., Aparicio, J. A. B., Rodríguez, J. J., Díez-Pastor, J.-F., García‐Osorio, C., & Davari, M. D. (2025) · Journal of Cheminformatics · doi:10.1186/s13321-025-01029-w · Cited by 2
“Another way of introducing the latent information from a large number of unlabelled sequences into the preprocessing step is described in [7], where the eUniRep protein sequence encoding method is presented (as an enhancement to the previously introduced UniRep [10]).”
Enhancing Antimicrobial Peptide Function Prediction via Knowledge Transfer on Protein Language Models
Liang, X., Zhao, H., & Wang, J. (2025) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3577565 · Cited by 3
“RecentprogressinNaturalLanguageProcessing(NLP)have ledtothedevelopmentofproteinlanguagemodels(pLMs) suchasEvolutionaryScaleModeling(ESM)[29],Uniﬁed Representation (UniRep) [30], and ProtTrans [31].”
Graph Neural Network-Based Approaches for Protein Function Prediction
Chaudhari, M., Bahmani, S., Pratyush, P., Garrett, S., Thapa, N. J., & KC, D. B. (2025) · Methods in molecular biology · doi:10.1007/978-1-0716-4662-5_3 · Cited by 2
Computational prediction of protein functional annotations
Kulmanov, M., & Hoehndorf, R. (2025) · Methods in molecular biology · doi:10.1007/978-1-0716-4662-5_1 · Cited by 2
Safe model based optimization balancing exploration and reliability for protein sequence design
Takizawa, S., Mori, K. P., Tanishiki, N., Yoshimura, D., Ohta, A., & Teramoto, R. (2025) · Scientific Reports · doi:10.1038/s41598-025-12568-5 · Cited by 3
Structure-Enhanced Protein Instruction Tuning: Towards General-Purpose Protein Understanding with LLMs
Wu, W., Wang, C., Chen, L., Yin, M., Zhu, Y., Fu, K., Ye, J., Xiong, H., & Wang, Z. (2025) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3711896.3737138 · Cited by 7
“In earlier works [4, 32, 75], LSTM and its variants [34, 37, 43, 100] were utilized to model the dependencies between residues in single protein sequences.”
Enhancing protein structural properties through model-guided sequence optimization
Ko, Y.-J., Kim, D., Muvva, C., Lee, W., Seo, M., & Park, K. (2025) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2025.147072 · Cited by 2
TCR representation learning with protein language models: a comprehensive review
Kinoshita, K., & Kobayashi, T. J. (2025) · International Immunology · doi:10.1093/intimm/dxaf048 · Cited by 3
“Notable examples include UniRep (37), TAPE (38), ESM2 (7), ProteinBERT (6), and ProGen (8), which are pre-trained on extensive protein databases like UniRef (39) and Pfam (40).”
HitScreen: A Sequence-Based Drug Virtual Screening Approach Using Data Augmentation and Protein Language Models
Chen, G., Liao, J., Yu, Y., Le, K., Zhao, H., Qin, Y., Cai, L., & Sheng, R. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c01753 · Cited by 4
Artificial intelligence for drug target identification
Li, Y., Huang, J., Sheng, Y., Wei, C., Liu, M., Wang, X., Chen, Y., Li, X., & Zheng, M. (2025) · Annual reports in medicinal chemistry · doi:10.1016/bs.armc.2025.08.002 · Cited by 2
Learnable protein representations in computational biology for predicting drug-target affinity
Kumar, R., Romano, J. D., & Ritchie, M. D. (2026) · Journal of Cheminformatics · doi:10.1186/s13321-025-01145-7 · Cited by 2
“Some generate representations by training a model to iteratively predict the next amino acid in a sequence such as UniRep [32] or by predicting masked parts of a sequence such as ESM Cambrian [33].”
Artificial Intelligence Platform EnzySFC for Enzyme Screening and Functional Conversion: Completely Redirecting Nitrilase to Nitrile Hydratase Function
Jiang, S., ZHENG, Z., Dong, H., Zhang, S., Tong, Q., & Yi, D. (2026) · Journal of Agricultural and Food Chemistry · doi:10.1021/acs.jafc.5c14148 · Cited by 2
Protein language models enable accurate viral host range prediction
Beltrán, J. F., Belén, L. H., Contreras, F. P., & Yáñez, A. J. (2026) · Scientific Reports · doi:10.1038/s41598-026-37765-8 · Cited by 2
DeepPFT-GANCap: An explainable attention-based hybrid deep learning framework integrating protein language models and wavelet denoising-based evolutionary features for photosystem-II core protein prediction
Alghamdi, W., Ali, F., Algarni, A., Alghushairy, O., Khalid, M., & Du, M. (2026) · Ecological Informatics · doi:10.1016/j.ecoinf.2026.103696 · Cited by 2
Machine learning for evolutionary genetics and molecular evolution
Svetec, N., Lee, U., & Zhao, L. (2026) · Trends in Genetics · doi:10.1016/j.tig.2026.01.013 · Cited by 2
Artificial intelligence-driven insights into the modulation of enzyme activity toward accelerated industrial applications
Han, H., Zhang, Q., Chen, Y., Wang, Z.-X., Su, X.-Y., Zhu, Y., Wang, X., & Feng, Z.-X. (2026) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2026.152018 · Cited by 2
Advances in computational and data-driven methodologies for accelerating antimicrobial peptide design and discovery
Jayawardena, A., Hung, A., Qiao, G. G., & Hajizadeh, E. (2026) · Progress in Polymer Science · doi:10.1016/j.progpolymsci.2026.102119 · Cited by 2
Machine Learning-Driven Simulations of the SARS-CoV-2 Fitness Landscape from Deep Mutational Scanning Experiments
Durumeric, A. E. P., McCarty, S., Smith, J., Köhler, J., Elez, K., Raich, L., Suriana, P., & Sztain, T. (2026) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.6c00332 · Cited by 2
“The final data set after alignment and filtering yielded 8,255,834 RBD sequences (91.55% of total), representing 27,980 unique RBD sequence variants consisting of combinations of 1311 distinct mutations.”
Influential
Analyzing Immunomes using Sequence Embedding and Network Analysis
Motuzenko, K., & Makarov, I. (2023) · IEEE World Symposium on Applied Machine Intelligence and Informatics (SAMI) · doi:10.1109/sami58000.2023.10044509 · Cited by 2
A review on the application of machine learning in production of woody biomass from natural and planted forests
Peng, W., & Sadaghiani, O. K. (2023) · Journal of Renewable and Sustainable Energy · doi:10.1063/5.0149793 · Cited by 2
PointListNet: Deep Learning on 3D Point Lists
Fan, H., Zhu, L., Yang, Y., & Kankanhalli, M. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.01697 · Cited by 2
SICGNN: structurally informed convolutional graph neural networks for protein classification
Lee, Y., Kim, E., Choi, J., & Lee, C.-H. (2024) · Machine Learning Science and Technology · doi:10.1088/2632-2153/ad979b · Cited by 2
“For example, these sequential strings can be tokenized into individual characters or fragments and then used for property prediction and classification of proteins or molecules through sequence-processing models such as recurrent neural networks or long short-term memory (LSTM) [6–8].”
Protein Structure–Function Relationship: A Kernel-PCA Approach for Reaction Coordinate Identification
Mollaei, P., & Farimani, A. B. (2025) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.5c00483 · Cited by 3
“The representation is also vital in protein engineering, 12 − 14 where the design of novel proteins with desired functions relies on a deep understanding of their structure − function relationships.”
A comparative study highlights superiority of LSTM in crop genomic prediction
Pan, R., Yang, Y., Zhang, Y., Xu, Q. J., Feng, Y., Chen, J., Li, W., He, S., Wei, X., & Zhang, M. (2025) · Planta · doi:10.1007/s00425-025-04843-6 · Cited by 3
“Notably, Alley et al. (2019) proposed UniRep, an LSTM-based model that distills essential protein features into compact statistical representations.”
Large-scale multi-omic biosequence transformers for modeling protein–nucleic acid interactions
Chen, S. F., Steele, R. J., Hocky, G. M., Lemeneh, B., Lad, S. P., & Oermann, E. K. (2026) · PLoS ONE · doi:10.1371/journal.pone.0341501 · Cited by 3
Inferring context-specific site variation with evotuned protein language models
Lytras, S., Strange, A. P., Ito, J., & Sato, K. (2026) · NAR Genomics and Bioinformatics · doi:10.1093/nargab/lqag018 · Cited by 3
“Instead of training new pLMs from scratch on different datasets, a so-lution proposed by Alley et al. (2019) is to fine-tune existing models to a local evolutionary context, referred to as “evotun-ing” [ 25 ].”
Sampling protein language models for functional protein design
Darmawan, J. T., Gal, Y., & Notin, P. (2026) · Cell Systems · doi:10.1016/j.cels.2026.101714 · Cited by 4
Irrational methods suggest indecomposability and emergence
Ghomi, H. T. (2023) · European Journal for Philosophy of Science · doi:10.1007/s13194-022-00503-8 · Cited by 2
Evo-velocity: Protein language modeling accelerates the study of evolution
Sandhu, M., Spence, M. A., & Jackson, C. J. (2022) · Cell Systems · doi:10.1016/j.cels.2022.03.004 · Cited by 2
Improved compound–protein interaction site and binding affinity prediction using self-supervised protein embeddings
Wu, J., Liu, Z., Yang, X., & Lin, Z. (2022) · BMC Bioinformatics · doi:10.1186/s12859-022-05107-w · Cited by 2
“UniRep [14], utilizing an mLSTM model pretrained on 24 million protein sequences, is one of the most effective self-supervised learning representations.”
Method
Experimental investigation of enzyme functional annotations reveals extensive annotation errorPreprint
Rembeza, E., & Engqvist, M. K. M. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.12.18.423474 · Cited by 2
“1) were downloaded and their full UniRep embeddings (Alley et al. 2019), of 5700 values, were computed.”
Method
A Deep Semi-Supervised Framework for Accurate Modelling of Orphan SequencesPreprint
Moffat, L., & Jones, D. T. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.07.13.201459 · Cited by 2
Self-Supervised Representation Learning of Protein Tertiary Structures (PtsRep) and Its Implications for Protein EngineeringPreprint
Luo, J., Cai, Y., Wu, J., Cai, H., Yang, X., & Lin, Z. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.12.22.423916 · Cited by 2
Construction of a neural network energy function for protein physicsPreprint
Yang, H., Xiong, Z., & Zonta, F. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.04.26.441401 · Cited by 2
Computational biology for enzymatic removal of mycotoxinsPreprint
Sandlin, N., Kish, D. R., Kim, J., Zaccaria, M., & Momeni, B. (2021) · doi:10.31219/osf.io/9f4an · Cited by 2
A study on experimental bias in post-translational modification predictorsPreprint
Zuallaert, J., Ramasamy, P., Bouwmeester, R., Callewaert, N., & Degroeve, S. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.11.28.518163 · Cited by 2
“PLMs are pretrained models that yield enriched, structure-aware sequence representations, instead of merely encoding the amino acid composition of a receptive field in a protein 29–35 .”
Background
Ultra-Accurate Classification and Discovery of Functional Protein-Coding Genes from Microbiomes Using FunGeneTyper: An Expandable Deep Learning-Based FrameworkPreprint
Zhang, G., Wang, H., Zhang, Z., Zhang, L., Guo, G., Yang, J., Yuan, F., & Ju, F. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.12.28.522150 · Cited by 2
Genetically encoded, noise-tolerant, auxin biosensors in yeast facilitate metabolic engineering and directed evolutionPreprint
Chaisupa, P., Rahman, M. M., Hildreth, S. B., Moseley, S., Gatling, C., Helm, R. F., & Wright, R. C. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.03.21.533585 · Cited by 2
Machine Learning to Predict Continuous Protein Properties from Simple Binary Sorting and Deep Sequencing DataPreprint
Case, M., Smith, M., Vinh, J., & Thurber, G. M. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.06.09.544229 · Cited by 2
Protein Structure Refinement via DeepTracer and AlphaFold2Preprint
Chen, J., Zia, A., Wang, F., Hou, J., Cao, R., & Si, D. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.08.16.553616 · Cited by 2
Zero-shot prediction of mutation effects on protein function with multimodal deep representation learningPreprint
Shu, W., Cheng, P., Mao, C., Tang, J., Yang, S., Gu, Q., Han, W., Yaofeng, C., Zhou, J., Li, W., et al. (2023) · Research Square · doi:10.21203/rs.3.rs-3358917/v1 · Cited by 2
Sequence, Structure and Functional space of Drosophila de novo proteinsPreprint
Middendorf, L., Iyengar, B. R., & Eicholt, L. A. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.01.30.577933 · Cited by 2
FS-mutant: A Few-shot Learning Benchmark for Protein Mutants MiningPreprint
Li, M., Yu, H., Fan, G., Zhou, Z., Tan, P., & Hong, L. (2024) · Research Square · doi:10.21203/rs.3.rs-3987561/v1 · Cited by 2
GeoAB: Towards Realistic Antibody Design and Reliable Affinity MaturationPreprint
Lin, H., Wu, L., Huang, Y., Liu, Y., Zhang, O., Zhou, Y., Sun, R., & Li, S. Z. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.05.15.594274 · Cited by 2
“Deep-learning-based methods concentrated on 1D sequence de-sign (Alley et al., 2019; Liu et al., 2020; Saka et al., 2021; Akbar et al., 2022), and then, thanks to great process in graph neural networks (Wu et al., 2021; Liu et al., 2021; Wu et al., 2023) and protein modeling (Huang et al., 2023; Wu…”
Method
Sable: Bridging the Gap in Protein Structure Understanding with an Empowering and Versatile Pre-training ParadigmPreprint
Ye, Q., li, J., Chen, X., Huang, H., Zeng, M., Yu, J., & Gong, X. (2024) · Research Square · doi:10.21203/rs.3.rs-4647798/v1 · Cited by 2
Phylogenetic Analysis and Machine Learning Identify Signatures of Selection and Predict Deleterious Mutations in Common BeanPreprint
Cordoba-Novoa, H. A., Buckler, E. S., Romay, M. C., Berthel, A., Johnson, L., Balasubramanian, P. M., & Hoyos‐Villegas, V. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.05.05.652309 · Cited by 2
GOLF: A Generative AI Framework for Pathogenicity Prediction of Myocilin OLF VariantsPreprint
Walton, T., Tsui, D., Fogel, L., Huard, D. J. E., Chagas, R. S., Lieberman, R. L., & Aghazadeh, A. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.06.17.660210 · Cited by 2
RemoteFoldSet: Benchmarking Structural Awareness of Protein Language ModelsPreprint
Ma, Z., & Bethel, N. P. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.09.23.678152 · Cited by 2
“For the benchmark, we include a representative selection of pLMs, comprising UniRep[13], multiple variants of ProGen2[14], CARP[15], several models from the ESM family[2], as well as the ProtT5 and ProtBert models from the ProtTrans suite[4].”
StoPred: Accurate Stoichiometry Prediction for Protein Complexes Using Protein Language Models and Graph AttentionPreprint
Liu, Q., Peng, C., Zheng, W. H., Zhang, C., & Freddolino, L. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.10.20.683515 · Cited by 2
“These models are capable of learning structural and functional features embedded within amino acid sequences, including secondary structure, cell localization, and co-evolutionary relationships [22–24].”
An AI-Native Biofoundry for Autonomous Enzyme Engineering: Integrating Active Learning with Automated ExperimentationPreprint
Zhang, C.-W., Yang, L., Qin, Y.-J., Li, D., Dong, S., & Yang, M. (2026) · bioRxiv · doi:10.64898/2026.02.01.703093 · Cited by 2
“Furthermore, navigating the vast, rugged protein fitness landscape typically relies on expert intuition or rational design, which is constrained by our incomplete understanding of complex structure-function relationships [8, 9].”
Deep models of protein evolution in time generate realistic evolutionary trajectories and functional proteinsPreprint
Koehl, A., Prillo, S., Liu, M., Xiong, J. T., Weng, L., Savage, D. F., & Song, Y. S. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.19.706898 · Cited by 5
“In a di ↵ erent vein, recent applications of deep learning have made great strides in probabilistic models of protein sequences [16–19].”
The rise of the machines in chemistry
Gill, M. L. (2022) · Magnetic Resonance in Chemistry · doi:10.1002/mrc.5304 · Cited by 1
Exploring artificial intelligence through a biologist's lens
Shabir, S., & Hamadani, A. (2024) · Elsevier eBooks · doi:10.1016/b978-0-443-24001-0.00001-4 · Cited by 1
AbDPP: Target‐oriented antibody design with pretraining and prior biological structure knowledge
Yu, C., Lin, X., Cheng, Y., Xu, J., Wang, H., Yan, Y., Huang, Y., Liu, L., Zhao, W., Zhao, Q., et al. (2024) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.26676 · Cited by 1
“Although current deep learning approaches for antibody sequence generation show promise, 20,23,25 there remains considerable room for improvement.”
Artificial intelligence, its components, and crucial technologies for its implementation
Bhattacharjee, R., Sasi, S. D., & Thakran, S. (2024) · Elsevier eBooks · doi:10.1016/b978-0-443-18508-3.00012-7 · Cited by 1
Progress in protein pre-training models integrating structural knowledge
Tang, T.-Y., Xiong, Y., Zhang, R.-G., Zhang, J., Li, W., Wang, J., & Wang, W. (2024) · Acta Physica Sinica · doi:10.7498/aps.73.20240811 · Cited by 2
Protein Sequence Feature Extraction Techniques: Research Advances, Progress, and Applications
Wan, X. (2025) · Biomedical engineering · doi:10.5772/intechopen.1009540 · Cited by 1
Shared-weight graph framework for comprehensive protein stability prediction across diverse mutation types
Li, G., Yao, S., & Fan, L. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf190 · Cited by 1
“Many models such as UniRep [20],TAPE [21],ProtTrans [22],and Evolutionary Scale Modeling (EMS) [23] demonstrate the power of deep learning, especially natural language processing, in zero-shot prediction and semantic representation of proteins.”
Progress and challenges for the application of machine learning for neglected tropical diseases
Khew, C., Akbar, R., & Mohd‐Assaad, N. (2025) · F1000Research · doi:10.12688/f1000research.129064.2 · Cited by 2
Multitask Learning-Based Approaches for Protein Function Prediction
Bahmani, S., Chaudhari, M., Carrier, C., Garrett, S., Pratyush, P., & KC, D. B. (2025) · Methods in molecular biology · doi:10.1007/978-1-0716-4662-5_4 · Cited by 1
Progress and challenges for the application of machine learning for neglected tropical diseases
Khew, C., Akbar, R., & Mohd‐Assaad, N. (2025) · F1000Research · doi:10.12688/f1000research.129064.3 · Cited by 1
Viral replication modulated by hallmark conformational ensembles: how AlphaFold-predicted features of RdRp folding dynamics combined with intrinsic disorder-mediated function enable RNA virus discovery
Tahzima, R., Charon, J., Díaz, A., Jonghe, K. D., Massart, S., Michon, T., & Vranken, W. (2025) · Frontiers in Virology · doi:10.3389/fviro.2025.1501616 · Cited by 1
Comparative analysis on artificial intelligence methods for DTI and DTBA prediction in drug repurposing
Kumar, S., & Singh, A. (2025) · Medicinal Chemistry Research · doi:10.1007/s00044-025-03465-7 · Cited by 1
“In contrast to the k-gram approach, a novel representation technique (referenced as [120]) was developed to use RNN using unlabelled amino acid sequences to construct statistical models of proteins.”
VF-Fuse: a dual-path feature fusion and iterative update architecture for virulence factor prediction
Huang, L., Yu, X., Li, S., Chen, Q., Xu, D., & Zhao, Q. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf481 · Cited by 1
“We also explored the Babbler model,part of the UniRep (Universal Representation of Protein) family [45].”
Design of a Labile RNase A Using Protein Language Models
Ong, G., Kong, K. W., Poh, S. E., Wong, F. T., Seow, Y., & Koh, W. L. C. (2025) · ACS Synthetic Biology · doi:10.1021/acssynbio.5c00287 · Cited by 2
Machine Learning Approaches for Interpretable Antibody Property Prediction Using Structural Data
Michalewicz, K., Barahona, M., & Bravi, B. (2026) · Methods in molecular biology · doi:10.1007/978-1-0716-5408-8_7 · Cited by 1
“Machine Learning (ML) methods hold great promise in this setting, thanks to their success in protein-related prediction tasks [9–11].”
LMProtein: a protein language model based framework for protein structural property prediction
Yuan, Y., Luo, H., & Tian, Y. (2025) · Physical Chemistry Chemical Physics · doi:10.1039/d5cp01861g · Cited by 1
KmPred: prediction of Michaelis constants (Km) using an integrative machine learning framework
Alazmi, M. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1711471 · Cited by 4
In silico discovery of umami peptides: From sequence-based intelligent models to structure–function mechanistic insight
Li, X., Luo, X., Xie, S., Wei, Y., Hong, F., Jiang, Y., Zhong, X., Xie, X., Ma, C., Hao, Y., et al. (2026) · Current Research in Food Science · doi:10.1016/j.crfs.2026.101404 · Cited by 1
“UniRep is architecturally based on a multiplicative LSTM (mLSTM) network (Alley et al., 2019) and was pre-trained on 24 million protein sequences from the UniRef50 database.”
Unraveling Glycation‐Induced Structure–Function Nexus in Food Proteins: From Analytical Innovations to AI‐Assisted Design
Chen, H., Tu, Z., Xie, Z., Wen, P., Wang, H., Wang, F., Zhang, J., Wang, S., & Hu, Y. (2026) · Comprehensive Reviews in Food Science and Food Safety · doi:10.1111/1541-4337.70477 · Cited by 1
Important progress in antimicrobial peptide prediction research in the past five years
Zuo, Y., Liu, X., Shao, X., Guo, Z., Zhu, S., & Dai, Z. (2026) · Analytical Biochemistry · doi:10.1016/j.ab.2026.116141 · Cited by 1
High-Throughput Screening and Mechanistic Elucidation of RhlA Mutants for Enhanced Rhamnolipid Biosynthesis Guided by EGCA-Net and Molecular Dynamics Simulations
王东培, Xu, C., Yang, Y., Jiang, Z., Wang, X., Zhao, Y., Li, R., & Ma, H. (2026) · Journal of Agricultural and Food Chemistry · doi:10.1021/acs.jafc.6c03371 · Cited by 1
AI-Driven Protein Research: From Prediction to Design
Min, G., & Peng, H. (2026) · Methods in molecular biology · doi:10.1007/978-1-0716-5249-7_13 · Cited by 1
Deep Learning and Artificial Intelligence in the Design of Angiotensin-Converting Enzyme (ACE) Inhibitory Peptides
Qian, Z., Zhou, Y., Lin, R., Hong, Y., Xu, Y., Li, S., Gao, J., Zhu, Q., & Zheng, H. (2026) · Journal of Artificial Intelligence Research and Innovation · doi:10.29328/journal.jairi.1001022 · Cited by 1
“The high-dimensional latent representations of deep generative models can capture the sequence semantics, structural patterns, and biochemical constraints of peptides, encoding the sequences into a continuous latent space for controlled exploration of functional peptide variants [60–62].”
PCNN-GP: A deep learning approach for globular protein prediction by integrating pulse-coupled neural network with protein language model
Abbas, M., Ali, F., Masmoudi, A., Alghamdi, W., Babour, A., & Asiry, O. (2026) · Journal of Computational Science · doi:10.1016/j.jocs.2026.102992 · Cited by 1
Generative AI designs functional thiolation domains for reprogramming non-ribosomal peptide synthetases
Bülbül, E. F., Bang, S., George, K., Bianchi, G., Raj, P., Chung, S., Pauline, V., Hochstrasser, R., Minas, H. A., Elgaher, W. A. M., et al. (2026) · Nature Communications · doi:10.1038/s41467-026-77963-6 · Cited by 1
“Detailed cloning procedures, expression and solubility protocols, puri ﬁ cation and refolding protocols, thermal stability measurements, and primer sequences are provided in the Supplementary Method 2.”

Authors: Alley, E. C., Khimulya, G., Biswas, S., AlQuraishi, M., & Church, G. M.
Affiliations: Harvard University (US)
Corresponding author: Church, George M.
ORCID: Alley, Ethan C.
Biswas, Surojit
AlQuraishi, Mohammed
Church, George M.
Venue: Nature Methods
Publisher: Springer Science and Business Media LLC
Published: 2019-10-21
Volume: 16
Issue: 12
Pages: 1315-1322
ISSN: 1548-7091 (ISSN-L)
1548-7105
DOI: 10.1038/s41592-019-0598-1
PMC: PMC7067682
OpenAlex: W2980789587
Cited by: 1,131
References: 80
Citations by year: 2026: 130 · 2025: 170 · 2024: 201 · 2023: 197 · 2022: 188 · 2021: 159 · 2020: 69 · 2019: 17
FWCI: 35.71
Type: journal article
Language: en
Open access: green
doi.org
Licence: cc-by-nc-nd
www.springer.com/tdm (tdm)
www.springer.com/tdm (vor)
Locations: Nature Methods (publishedVersion)
bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
Research Square (acceptedVersion, cc-by)
PubMed (publishedVersion)
Europe PMC (PubMed Central) (acceptedVersion)
PubMed Central (acceptedVersion)
OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
Funders: Amazon Web Services Cloud Credits for Research
Center for Effective Altruism
U.S. Department of Health & Human Services | National Institutes of Health (T32HG002295, U54-CA225088)
National Science Foundation GRFP Fellowship
U.S. Department of Health & Human Services | NIH | National Institute of General Medical Sciences (P50GM107618)
Topics: Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › RNA and protein synthesis mechanisms
Keywords: rational protein design, protein stability prediction, de novo protein design, protein mutants, recurrent neural networks, protein function prediction
MeSH terms: Deep Learning, Amino Acid Sequence, Mutation, Protein Engineering, Protein Stability

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

Unified rational protein engineering with sequence-based deep representation learning

Abstract

Overview

Integrity signals

Verification

Claims the paper makes

Claims other papers make about this one

Self-check

Citations

Cites

Cited by

Related

Metadata

Record

About these checks