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 accessPublished version of a preprint
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.
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?
The cited source says what the claim attributes to it.
In the citing paper
“They also learn summary representations, powerful features that can be used to better capture sequence relationships and link sequence to function via transfer learning 27,28,32-35.”
From Learning the protein language: Evolution, structure, and function (2021) · cites this paper as [28]
In the citing paper
“Since then, various language model-based protein embedding methods have been applied to these and other protein prediction problems through transfer learning, including protein phenotype prediction 28,32-34, residue-residue contact prediction 32,70, fold recognition 33, protein-protein 71,72 and protein-drug interaction prediction 35,73.”
From Learning the protein language: Evolution, structure, and function (2021) · cites this paper as [28]
In the citing paper
“In recent years, deep learning architectures that were originally developed for natural language processing (NLP) tasks, such as translating a sentence from one language into another, have been applied successfully to the creation of numerical protein representations from amino acid sequences 17 , 29 .”
From Turnover number predictions for kinetically uncharacterized enzymes using machine and deep learning (2023) · cites this paper as [29]
In the citing paper
“These fingerprints are used together with a 1900-dimensioanl enzyme representation, called UniRep vector 29 , as the input for a gradient boosted decision tree model 34 to predict the K M value for an enzyme-substrate pair.”
From Turnover number predictions for kinetically uncharacterized enzymes using machine and deep learning (2023) · cites this paper as [29]
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
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.”
MethodBerman, 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(2008) · Journal of Machine Learning Research · Cited by 33,051
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
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
Cited by
Sorted by citations
Items 601–700 of 1,003
Construction of a Deep Neural Network Energy Function for Protein Physics
Yang, H., Xiong, Z., & Zonta, F. (2022) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.2c00069 · Cited by 13
A structure-based deep learning framework for protein engineeringPreprint
Shroff, R., Cole, A. W., Morrow, B. R., Diaz, D. J., Donnell, I., Gollihar, J., Ellington, A. D., & Thyer, R. T. (2019) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/833905 · Cited by 10
Reimplementing Unirep in JAXPreprint
Ma, E. J., & Kummer, A. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.05.11.088344 · Cited by 12
Neural Potts ModelPreprint
Sercu, T., Verkuil, R., Meier, J., Amos, B., Lin, Z., Chen, C., Liu, J. J., LeCun, Y., & Rives, A. W. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.04.08.439084 · Cited by 10
“Alley et al. (2019), Heinzinger et al. (2019), and Madani et al. (2020) trained models with autoregressive objectives.”
BackgroundPre-training Co-evolutionary Protein Representation via A Pairwise Masked Language ModelPreprint
He, L., Zhang, S., Wu, L., Xia, H., Ju, F., Zhang, H., Liu, S., Xia, Y., Zhu, J., Deng, P., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.15527 · Cited by 35
“UniRep (Alley et al., 2019) uses a similar training scheme as Bepler & Berger (2019), then use evo-turning (evolutionary fine-tuning) to address the importance of evolutionary information for the protein embedding.”
MethodIllig, A.-M., Siedhoff, N. E., Schwaneberg, U., & Davari, M. D. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.06.07.495081 · Cited by 10
Systematic multi-trait AAV capsid engineering for efficient gene deliveryPreprint
Eid, F.-E., Chen, A. T., Chan, K. Y., Huang, Q., Zheng, Q., Tobey, I. G., Pacouret, S., Brauer, P. P., Keyes, C., Powell, M. M., et al. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.12.22.521680 · Cited by 10
Review of predicting protein stability changes upon variations
Qiu, Y., Huang, T., & Cai, Y. (2024) · PROTEOMICS · doi:10.1002/pmic.202300371 · Cited by 10
Computational tools for plant genomics and breeding
Wang, H., Chen, M., Wei, X., Xia, R., Pei, D., Huang, X., & Han, B. (2024) · Science China Life Sciences · doi:10.1007/s11427-024-2578-6 · Cited by 9
“For example, UniRep was trained to perform the next amino-acid prediction and learn how to represent proteins as a fixed-length vector (Alley et al., 2019).”
Tzavella, K., Díaz, A., Olsen, C., & Vranken, W. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae664 · Cited by 9
Predicting purification process fit of monoclonal antibodies using machine learning
Maier, A. J., Cha, M., Burgess, S., Wang, A., Cuellar, C., Kim, S. Y., Rajan, N. S., Neyyan, J., SENGUPTA, R., O’Connor, K., et al. (2025) · mAbs · doi:10.1080/19420862.2024.2439988 · Cited by 11
Unveiling the ghost: machine learning’s impact on the landscape of virology
Bowyer, S., Allen, D. J., & Furnham, N. (2025) · Journal of General Virology · doi:10.1099/jgv.0.002067 · Cited by 13
“One recent model, known as multitask transfer, is based on an LLM named UniRep, which is trained on over 24 million sequences from UniRef50 [92, 99, 100].”
Prediction of Adeno-Associated Virus Fitness with a Protein Language-Based Machine Learning Model
Wu, J., Qiu, Y., Lyashenko, E., Torregrosa, T., Pfister, E. L., Ryan, M. J., Mueller, C., & Choudhury, S. (2025) · Human Gene Therapy · doi:10.1089/hum.2024.227 · Cited by 10
“6 – 9 PLMs treat protein sequences as sentences in which amino acid residues represent words.”
DrugTar improves druggability prediction by integrating large language models and gene ontologies
Borhani, N., Izadi, I., Motahharynia, A., Sheikholeslami, M., & Gheisari, Y. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf360 · Cited by 14
“Finally, a sigmoid activation function is applied to the output neuron to predict a continuous druggability score for the protein, ranging from zero to one.”
Locality-aware pooling enhances protein language model performance across varied applications
Hoang, M., & Singh, M. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf178 · Cited by 10
“Protein language models (PLMs) have recently emerged as a new technique for modeling the biological “grammar” that governs the behaviors of proteins (Alley et al. 2019, Bepler and Berger 2019, Heinzinger et al. 2019, Rao et al. 2019, Bepler and Berger 2021, Elnaggar et al. 2021, Rives et al. 2021,…”
An automated multi-modal graph-based pipeline for mouse genetic discovery
Fang, Z., & Peltz, G. (2022) · Bioinformatics · doi:10.1093/bioinformatics/btac356 · Cited by 7
Connecting the Dots in Self-Supervised Learning: A Brief Survey for Beginners
Fang, P., Li, X., Yang, Y., Zhang, S., Kang, Q.-Y., Li, X., & Lan, Z. (2022) · Journal of Computer Science and Technology · doi:10.1007/s11390-022-2158-x · Cited by 9
“UniRep [73] takes a single amino acid as a word and uses the multiplicative LSTM [100] to train a generative model in an auto-regressive manner [101] .”
BackgroundMethodInfluentialGenerating new protein sequences by using dense network and attention mechanism
Wang, F., Feng, X., Kong, R., & Chang, S. (2022) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2023195 · Cited by 9
Peng, W., & Sadaghiani, O. K. (2023) · International Journal of Green Energy · doi:10.1080/15435075.2023.2255647 · Cited by 10
Anteghini, M., dos Santos, V. A. P. M., & Saccenti, E. (2023) · Journal of Cellular Biochemistry · doi:10.1002/jcb.30490 · Cited by 9
“…methods for the embedding of protein sequences based on deep ‐ learning approaches and protein sequences: The unified representation ( UniRep ) 20 : Is based on a 1900 ‐ hidden unit recurrent neural network architecture, able to capture evolutionary, chemical and biological information…”
MethodGrowing ecosystem of deep learning methods for modeling protein–protein interactions
Rogers, J. R., Nikolényi, G., & AlQuraishi, M. (2023) · Protein Engineering Design and Selection · doi:10.1093/protein/gzad023 · Cited by 12
Self-citationDeep-Representation-Learning-Based Classification Strategy for Anticancer Peptides
Khan, S. (2024) · Mathematics · doi:10.3390/math12091330 · Cited by 9
Ensemble Machine Learning and Predicted Properties Promote Antimicrobial Peptide Identification
Zhong, G., Liu, H., & Deng, L. (2024) · Interdisciplinary Sciences Computational Life Sciences · doi:10.1007/s12539-024-00640-z · Cited by 9
“The sequential features are derived from three pre-trained embedding approaches, including soft symmetric alignment (SSA) [31], unified representation (UniRep) [32], and ESM-1b [33].”
In silico evolution of nucleic acid-binding proteins from a nonfunctional scaffold
Raven, S. A., Payne, B., Bruce, M., Filipovska, A., & Rackham, O. (2022) · Nature Chemical Biology · doi:10.1038/s41589-022-00967-y · Cited by 9
“These studies take mutations identified in laboratory directed evolution experiments to build sequence–function models and explore further sequence space 34–39 .”
BackgroundTransfer 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
Turnover number predictions for kinetically uncharacterized enzymes using machine and deep learningPreprint
Kroll, A., Hu, X.-P., Liebrand, N. A., & Lercher, M. J. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.11.10.516024 · Cited by 9
SEQUENCE 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
“DL-based PFP methods include function prediction from AA-sequence (Rao et al., 2019; Alley et al., 2019; Elnaggar et al., 2020; Dallago et al., 2021; Kulmanov & Hoehndorf, 2020; Meier et al., 2021; Biswas et al., 2021; Gelman et al., 2021; Yang et al., 2022a), 3-dimensional structure (Gligorijevi´c…”
InfluentialSun, J., Zhu, T., Cui, Y., & Wu, B. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.08.09.552725 · Cited by 9
“MLM has found widespread application across protein sequences 28 , MSAs 29 , and protein structures 30 , especially in the prediction of mutation fitness.”
BackgroundWang, Y., Wang, Z., Sadeh, G., Zancato, L., Achille, A., Karypis, G., & Rangwala, H. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.10.29.620988 · Cited by 12
“To train LMFold, we use the Frame Aligned Point Error (FAPE) and distogram losses introduced in AlphaFold2 (Jumper et al., 2021), as well as heads for predicting LDDT and the pTM score .”
Investigating the determinants of performance in machine learning for protein fitness prediction
Sandhu, M., Mater, A. C., Matthews, D. S., Spence, M. A., Lenskiy, A. A., & Jackson, C. (2020) · doi:10.1101/2020.09.30.319780 · Cited by 8
Kandathil, S. M., Greener, J. G., Lau, A. M., & Jones, D. T. (2020) · doi:10.1101/2020.11.27.401232 · Cited by 8
“Standard transformer models have already been used to embed unaligned protein sequences [39–41], but the most efficient transformer models released in the last year are now capable of handling sequence lengths even in the millions, and so this suggests that a single deep transformer model with a compressed self-atte…”
MethodMalik, M. S., Chang, Y., Liu, Y., Le, V. T., & Ou, Y. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c00961 · Cited by 8
Yang, S., Ni, J., & Xu, P. (2024) · Journal of Agricultural and Food Chemistry · doi:10.1021/acs.jafc.4c05650 · Cited by 8
A robust ensemble framework for anticancer peptide classification using multi-model voting approach
Abbas, Z., Kim, S., Lee, N., Kazmi, S. I. H., & Lee, S. W. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.109750 · Cited by 8
Shoombuatong, W., Mookdarsanit, P., Mookdarsanit, L., Schaduangrat, N., Ahmed, S., Kabir, M., & Chumnanpuen, P. (2025) · Interdisciplinary Sciences Computational Life Sciences · doi:10.1007/s12539-025-00696-5 · Cited by 8
“To date, several pLMs have been constructed based on varying model sizes, architectures, and pretraining datasets, such as ProtTrans [24], UniRep [22], and ESM [27, 28].”
InfluentialAhmed, Z., Shahzadi, K., Li, R., Jiang, Y., Jin, Y.-T., Arif, M., & Feng, J. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf313 · Cited by 8
LassoESM a tailored language model for enhanced lasso peptide property prediction
Mi, X., Barrett, S. E., Mitchell, D. A., & Shukla, D. (2025) · Nature Communications · doi:10.1038/s41467-025-63412-3 · Cited by 7
Artificial intelligence‐enabled digital biomedical engineering
Song, P., Tang, X., Lv, X., Reis, R. L., Chen, X., Bai, L., & Su, J. (2025) · BMEMat · doi:10.1002/bmm2.70049 · Cited by 9
Golinski, A. W., Schmitz, Z. D., Nielsen, G. H., Johnson, B., Saha, D., Appiah, S., Hackel, B. J., & Martiniani, S. (2023) · ACS Synthetic Biology · doi:10.1021/acssynbio.3c00196 · Cited by 7
Classifying alkaliphilic proteins using embeddings from protein language model
Susanty, M., Mursalim, M. K. N., Hertadi, R., Purwarianti, A., & Mengko, T. L. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108385 · Cited by 10
Yang, Q., Jin, X., Zhou, H., Ying, J., Zou, J., Liao, Y., Lu, X., Ge, S., Yu, H. J., & Min, X. (2024) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2024.108067 · Cited by 8
Comprehensive Research on Druggable Proteins: From PSSM to Pre-Trained Language Models
Chu, H., & Liu, T. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms25084507 · Cited by 8
“Among all sorts of language models, the ESM-2 model class was chosen for its speed and effectiveness [19].”
MethodSCLpred-ECL: Subcellular Localization Prediction by Deep N-to-1 Convolutional Neural Networks
Gillani, M., & Pollastri, G. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms25105440 · Cited by 8
“Another example is OrganelX [27], a hybrid predictor that uses two different approaches, i.e., protein sequence embedding via Unified Representation (UniRep) [28] and Sequence-to-Vector (SeqVec) [29].”
MethodLi, Y., Wang, C., Gu, H., Feng, H., & Ruan, Y. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad761c · Cited by 9
Fast and Functional Structured Data Generators Rooted in Out-of-Equilibrium Physics
Carbone, A., Decelle, A., Rosset, L., & Seoane, B. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3495999 · Cited by 10
Machine learning approaches for predicting protein-ligand binding sites from sequence data
Vural, O., & Jololian, L. (2025) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2025.1520382 · Cited by 8
Tamaki, F. K. (2020) · Emerging Topics in Life Sciences · doi:10.1042/etls20200047 · Cited by 8
Prediction of antibody binding to SARS-CoV-2 RBDs
Wang, E. J. D. (2022) · Bioinformatics Advances · doi:10.1093/bioadv/vbac103 · Cited by 9
Cross-Gate MLP with Protein Complex Invariant Embedding Is a One-Shot Antibody Designer
Tan, C., Gao, Z., Wu, L., Xia, J., Zheng, J., Yang, X., Liu, Y., Hu, B., & Li, S. Z. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i14.29445 · Cited by 11
Deep learning enables the design of functional de novo antimicrobial proteinsPreprint
Caceres‐Delpiano, J., Ibañez, R., Alegre, P., Sanhueza, C., Paz-Fiblas, R., Correa, S., Retamal, P. A., Jiménez, J. C., & Álvarez, L. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.08.26.266940 · 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
Targeting AAV vectors to the CNS via de novo engineered capsid-receptor interactionsPreprint
Huang, Q., Chen, A. T., Chan, K. Y., Sorensen, H., Barry, A. J., Azari, B., Beddow, T., Zheng, Q., Zhao, B., Tobey, I. G., et al. (2022) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2022.10.31.514553 · Cited by 8
“While our SVAE uses a standard one-hot encoding scheme which worked well using the high-quality data from the pull-down assays, others have experimented with the use of other encoding schemes such as physicochemical parameters (Georgiev, 2009) or learned representations (Alley et al., 2019).”
MethodRamon, A., Ali, M., Atkinson, M., Saturnino, A., Didi, K., Visentin, C., Ricagno, S., Xu, X., Greenig, M., & Sormanni, P. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.04.28.538712 · Cited by 8
“The embedding of the VQ-VAE, and possibly also the last hidden layer of the decoder, can be seen as an effective way to distil the fundamental features of antibody variable domains into a representation that is semantically rich and structurally, evolutionarily, and biophysically grounded (63).”
BackgroundRapid protein evolution by few-shot learning with a protein language modelPreprint
Jiang, K.-Y., Yan, Z.-Q., Bernardo, M. D., Sgrizzi, S. R., Villiger, L., Kayabolen, A., Kim, B., Carscadden, J. K., Hiraizumi, M., Nishimasu, H., et al. (2024) · bioRxiv · doi:10.1101/2024.07.17.604015 · Cited by 9
Safeguarding Mail-Order DNA Synthesis in the Age of Artificial Intelligence
Batalis, S., Schuerger, C., Grönvall, G. K., & Walsh, M. (2023) · Applied Biosafety · doi:10.1089/apb.2023.0020 · Cited by 7
Seq-RBPPred: Predicting RNA-Binding Proteins from Sequence
Yan, Y., Li, W., Wang, S., & Huang, T. (2024) · ACS Omega · doi:10.1021/acsomega.3c08381 · Cited by 8
“To accomplish this, our approach involves the utilization of Protr, 30 UniRep, 31 SeqVec, 32 and ESM-1b 33 to extract a comprehensive set of 6944 features from protein sequences.”
MethodExcelzyme: A Swiss University-Industry Collaboration for Accelerated Biocatalyst Development
Malca, S. H., Stockinger, P., Duss, N., Milbredt, D., Iding, H., & Buller, R. (2024) · CHIMIA International Journal for Chemistry · doi:10.2533/chimia.2024.108 · Cited by 7
ACP-ESM2: The prediction of anticancer peptides based on pre-trained classifier
Song, H., Lin, X., Zhang, H., & Yin, H. (2024) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2024.108091 · Cited by 7
Changiarath, A., Arya, A., Xenidis, V. A., Padeken, J., & Stelzl, L. S. (2024) · Faraday Discussions · doi:10.1039/d4fd00099d · Cited by 7
Bioengineered therapeutic systems for improving antitumor immunity
Cao, Y., Yan, W., Yi, W., Yin, Q., & Li, Y. (2024) · National Science Review · doi:10.1093/nsr/nwae404 · Cited by 8
“In addition to assisting in modifying strategies, AI can also enhance the efficiency of cell engineering toolkits such as gene editing enzymes through directed evolution [142 ,143 ].”
Li, G., Zhang, N., & Fan, L. (2025) · ACS Omega · doi:10.1021/acsomega.4c09688 · Cited by 7
Zhang, N., Deng, Y., Li, R., Liu, F., Wei, D., Ren, Z., Jin, X., Zhao, T., Park, M., Choi, J., et al. (2025) · Free Radical Biology and Medicine · doi:10.1016/j.freeradbiomed.2025.08.015 · Cited by 7
Structure-informed machine learning for drug discovery: a task-centric perspective
Li, Y., Zhan, R.-H., Rao, J., Liu, M., Sang, P., Zeng, X., Zheng, M., Li, X., & Yang, L. (2026) · Briefings in Bioinformatics · doi:10.1093/bib/bbag081 · Cited by 8
“On the other hand, sequence-based learning frameworks (e.g. compound–protein interaction prediction) lack spatial resolution and fail to capture the geometric and physicochemical complexity of binding interactions [9, 10].”
Wagner, A. (2024) · Bioinformatics · doi:10.1093/bioinformatics/btae317 · Cited by 7
“Several studies have used machine learning to predict molecular phenotypes that can be correlated with fitness (Alley et al. 2019, Li et al. 2019, Xu et al. 2020, Wittmann et al. 2021, Tareen et al. 2022, Vaishnav et al. 2022).”
BackgroundSusanty, M., Mursalim, M. K. N., Hertadi, R., Purwarianti, A., & Rajab, T. L. (2024) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2024.108163 · Cited by 10
Acercando la inteligencia artificial a los servicios de farmacia hospitalaria
González-Pérez, Y., Delgado, A. M., & Sesmero, J. M. M. (2024) · Farmacia Hospitalaria · doi:10.1016/j.farma.2024.02.007 · Cited by 16
Wang, B., & Li, W. (2024) · Genes · doi:10.3390/genes15081090 · Cited by 7
A protein fitness predictive framework based on feature combination and intelligent searching
Zhang, Z., Li, Z., Wang, Q., Wu, H., Yang, M., Zhao, F., Tan, M., & Han, S. (2024) · Protein Science · doi:10.1002/pro.5211 · Cited by 7
How did we get there? AI applications to biological networks and sequences
Anteghini, M., Gualdi, F., & Oliva, B. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110064 · Cited by 9
Improving sequence-based modeling of protein families using secondary-structure quality assessment
Malbranke, C., Bikard, D., Cocco, S., & Monasson, R. (2021) · Bioinformatics · doi:10.1093/bioinformatics/btab442 · Cited by 8
van den Bent, I., Makrodimitris, S., & Reinders, M. J. T. (2021) · Evolutionary Bioinformatics · doi:10.1177/11769343211062608 · Cited by 9
Anteghini, M., Haja, A., dos Santos, V. A. P. M., Schomaker, L., & Saccenti, E. (2022) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2022.11.058 · Cited by 8
“The Unified Representation (UniRep) [9] and the Sequence-to-Vector (SeqVec) [10] are two of the most promising and already used protein sequence embeddings.”
InfluentialVideo frame prediction of microbial growth with a recurrent neural network
Robertson, C., Wilmoth, J. L., Retterer, S. T., & Fuentes‐Cabrera, M. (2023) · Frontiers in Microbiology · doi:10.3389/fmicb.2022.1034586 · Cited by 7
“RNNs have recently been used to study proteins (Alley et al., 2019; Le, 2019; Tng et al., 2022), whole bacteria genomes (Alley et al.”
BackgroundMethodEmpowering Protein Engineering through Recombination of Beneficial Substitutions
Wang, X., Li, A., Li, X., & Cui, H. (2024) · Chemistry - A European Journal · doi:10.1002/chem.202303889 · Cited by 7
Universal Transforming Geometric NetworkPreprint
Li, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.00723 · Cited by 10
Evaluation of Methods for Protein Representation Learning: A Quantitative AnalysisPreprint
Ünsal, S., Ataş, H., Albayrak, M., Turhan, K., Acar, A. C., & Doğan, T. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.10.28.359828 · Cited by 7
fastMSA: Accelerating Multiple Sequence Alignment with Dense Retrieval on Protein LanguagePreprint
Hong, L., Sun, S., Zheng, L., Tan, Q., & Li, Y. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.12.20.473431 · Cited by 9
When Geometric Deep Learning Meets Pretrained Protein Language ModelsPreprint
Wu, F., Tao, Y., Radev, D., & Xu, J. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.01.05.522958 · Cited by 7
Learning Complete Protein Representation by Deep Coupling of Sequence and StructurePreprint
Hu, B., Tan, C., Xia, J., Zheng, J., Huang, Y., Wu, L., Liu, Y., Xu, Y., & Li, S. Z. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.07.05.547769 · Cited by 7
“As proteins are sequences of amino acids, considering their similarities with human languages, UniRep [1], UDSMProt [42] and SeqVec [23] use LSTM or its variants to learn sequence representations and long-range dependencies.”
MethodEvaluating variant effect prediction across virusesPreprint
Gurev, S. F., Youssef, N., Jain, N., Mehrotra, A., Leung, S. R. M., Jackson, A., & Marks, D. S. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.08.04.668549 · Cited by 10
Predicting antibiotic resistance genes and bacterial phenotypes based on protein language models
Wang, B., Meng, R., Li, Z., Hu, M., Wang, X., Zhao, Y., Chai, Z., Jin, Y., Yue, J., Chen, W., et al. (2025) · Frontiers in Microbiology · doi:10.3389/fmicb.2025.1628952 · Cited by 9
Chisompola, D., Luwaya, E. L., Nzobokela, J., Mwansa, P., & Chakulya, M. (2026) · Frontiers in Microbiology · doi:10.3389/fmicb.2025.1717859 · Cited by 10
“By applying learnable filters across nucleotide or amino acid sequences, CNNs act as motif discovery engines, identifying conserved short patterns such as transcription factor binding sites or protease cleavage motifs (Alley et al., 2019; Jumper et al., 2021; Senior et al., 2020).”
Advanced molecular approaches in male infertility diagnosis
Botezatu, A., Vladoiu, S. V., Fudulu, A., Albulescu, A., Plesa, A., Muresan, A., Stancu, C., Iancu, I. V., Diaconu, C. C., Velicu, A., et al. (2022) · Biology of Reproduction · doi:10.1093/biolre/ioac105 · Cited by 7
A sweeter future: Using protein language models for exploring sweeter brazzein homologs
Chua, B. N., Guo, W. M., Wong, H. T., Ow, D. S., Ho, P. L., Koh, W. L. C., Koay, A. Z., & Wong, F. T. (2023) · Food Chemistry · doi:10.1016/j.foodchem.2023.136580 · Cited by 6
Qu, Y., Niu, Z., Ding, Q., Zhao, T., Kong, T., Bai, B., Ma, J., Zhao, Y., & Zheng, J. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms242216496 · Cited by 7
“In the realm of deep learning, various disciplines are advancing towards the development of “large models” [13].”
BackgroundMethodInfluentialExploiting Hierarchical Interactions for Protein Surface Learning
Lin, Y., Pan, L., Li, Y., Liu, Z., & Li, X. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3356231 · Cited by 5
“Therefore, exploring the functions of different proteins’ 3D structures is crucial for understanding their working mechanism, which benefits many applications, such as new drug development [1].”
BackgroundExploring the Knowledge of an Outstanding Protein to Protein Interaction Transformer
Yang, S., Cheng, P., Liu, Y., Feng, D., & Wang, S. (2024) · IEEE/ACM Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2024.3381825 · Cited by 6
“Unirep [37] explores the pre-training and finetuning paradigm.”
BackgroundLiang, X., Zhao, H., & Wang, J. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-5131-0_6 · Cited by 6
Antibody-SGM, a Score-Based Generative Model for Antibody Heavy-Chain Design
Xie, X., Valiente, P. A., Lee, J. S., Kim, J., & Kim, P. M. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c00711 · Cited by 7
learnMSA2: deep protein multiple alignments with large language and hidden Markov models
Becker, F., & Stanke, M. (2024) · Bioinformatics · doi:10.1093/bioinformatics/btae381 · Cited by 6
Zhang, W., Ding, Y., Wei, L., Guo, X., & Ni, F. (2024) · Briefings in Bioinformatics · doi:10.1093/bib/bbae534 · Cited by 6
“Nowadays, machine learning algorithms have been combined with feature vectors obtained using deep learning feature extraction techniques for therapeutic peptide prediction [18–22].”
BackgroundAbdollahi, S., Schaub, D. P., Barroso, M., Laubach, N. C., Hutwelker, W., Panzer, U., Gersting, S. W., & Bonn, S. (2024) · Journal of Cheminformatics · doi:10.1186/s13321-024-00913-1 · Cited by 6
“UniRep-based features : The UniRep-based input features are derived from a multiplicative long short-term memory (mLSTM) model [36], which has been trained on a vast dataset comprising approximately 25 million amino acid sequences from the UniRef50 clusters of Uni-Prot.”
MethodRepeat proteins: designing new shapes and functions for solenoid folds
Gidley, F., & Parmeggiani, F. (2021) · Current Opinion in Structural Biology · doi:10.1016/j.sbi.2021.02.002 · Cited by 6
Narayanan, P. S., & Runthala, A. (2022) · Biocatalysis and Biotransformation · doi:10.1080/10242422.2022.2030317 · Cited by 6
“For extracting the underlying functionally crucial features of a protein, Ethan’s group has deployed deep learning methodology in their methodologies for explaining the structural, biophysical and evolutionary relationships within the protein sequences [45].”
BackgroundMethodInfluentialSmart de novo Macromolecular Structure Modeling from Cryo-EM Maps
Si, D., Chen, J., Nakamura, A., Chang, L., & Guan, H. (2023) · Journal of Molecular Biology · doi:10.1016/j.jmb.2023.167967 · Cited by 6
González, M. G., van den Broek, R. L., Braun, T. G. M., Chatzopoulou, M., Jespers, W., IJzerman, A. P., Heitman, L. H., & van Westen, G. J. P. (2023) · Journal of Cheminformatics · doi:10.1186/s13321-023-00745-5 · Cited by 6
“As protein embeddings Unirep [60] were used, pre-calculated in the Papyrus dataset.”
MethodFu, Y., Bedő, J., Papenfuss, A. T., & Rubin, A. F. (2022) · GigaScience · doi:10.1093/gigascience/giad073 · Cited by 5
STAR: A Web Server for Assisting Directed Protein Evolution with Machine Learning
Yang, L., Liang, X., Zhang, N., & Lu, L. (2023) · ACS Omega · doi:10.1021/acsomega.3c04832 · Cited by 6
Antibody complementarity-determining region design using AlphaFold2 and DDG predictor
Ueki, T., & Ohue, M. (2024) · The Journal of Supercomputing · doi:10.1007/s11227-023-05887-9 · Cited by 7
Metadata
- Authors
- Alley, E. C., Khimulya, G., Biswas, S., AlQuraishi, M., & Church, G. M.
- Affiliations
- Harvard University (US)
- Corresponding author
- Church, George M.
- Venue
- Nature Methods
- Published
- 2019-10-21
- Volume
- 16
- Issue
- 12
- Pages
- 1315-1322
- ISSN
- 1548-7091 (ISSN-L)
- 1548-7105
- 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
- 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
