Generating novel protein sequences using Gibbs sampling of masked language models
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
Recently developed language models (LMs) based on deep neural networks have demonstrated the ability to generate fluent natural language text. LMs pre-trained on protein sequences have shown state of the art performance on a variety of downstream tasks. Protein LMs have also been used to generate novel protein sequences. In the present work we use Gibbs sampling of BERT-style LMs, pre-trained on protein sequences using the masked language modeling task, to generate novel protein sequences. We evaluate the quality of the generated sequences by comparing them to natural sequences from the same family. In particular, we focus on proteins from the chorismate mutase type II family, which has been used in previous work as an example target for protein generative models. We find that the Gibbs sampling process on BERT-style models pretrained on millions to billions of protein sequences is able to generate novel sequences that retain key features of related natural sequences. Further, we find that smaller models fine-tuned or trained from scratch on family-specific data are able to equal or surpass the generation quality of large pre-trained models by some metrics. The ability to generate novel natural-like protein sequences could contribute to the development of improved protein therapeutics and protein-catalysts for industrial chemical production.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821
Rives, A., Meier, J., Sercu, T., Goyal, S., Lin, Z., Liu, J., Guo, D., Ott, M., Zitnick, C. L., Ma, J., et al. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2016239118 · Cited by 3,286
Rives, A., Meier, J., Sercu, T., Goyal, S., Lin, Z., Liu, J., Guo, D., Ott, M., Zitnick, C. L., Ma, J., et al. (2019) · doi:10.1101/622803 · Cited by 3,280 · cited 18×
“Fine-tuning only the last layer of ESM34 or ProtBERT might also improve performance on a target family without needing significant compute power for the back propagation.”
BackgroundMethodInfluentialImproved protein structure prediction using predicted interresidue orientations
Yang, J., Anishchenko, I., Park, H., Peng, Z., Ovchinnikov, S., & Baker, D. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1914677117 · Cited by 1,421
Unified rational protein engineering with sequence-based deep representation learning
Alley, E. C., Khimulya, G., Biswas, S., AlQuraishi, M., & Church, G. M. (2019) · Nature Methods · doi:10.1038/s41592-019-0598-1 · Cited by 1,131 · cited 2×
“…is an RNN model trained on a subset of UniProt; one key finding from the UniRep work is that using the pre-training task (in their case, next token prediction) to fine-tune on additional examples of proteins from a particular family, increases performance on down-stream tasks for that family [1, 5].”
MethodEvaluating Protein Transfer Learning with TAPE
Rao, R., Bhattacharya, N., Thomas, N., Duan, Y., Chen, X., Canny, J., Abbeel, P., & Song, Y. S. (2019) · doi:10.1101/676825 · Cited by 719
Protein-level assembly increases protein sequence recovery from metagenomic samples manyfold
Steinegger, M., Mirdita, M., & Söding, J. (2019) · Nature Methods · doi:10.1038/s41592-019-0437-4 · Cited by 445 · cited 8×
“Pre-trained ESM12 produced sequences of much poorer quality than ESM34 and ProtBERT-BFD, but fine-tuning ESM12 allowed generation of sequences whose quality met or surpassed the quality of sequences generated from larger models.”
BackgroundInfluentialLow-N protein engineering with data-efficient deep learning
Biswas, S., Khimulya, G., Alley, E. C., Esvelt, K. M., & Church, G. M. (2021) · Nature Methods · doi:10.1038/s41592-021-01100-y · Cited by 422 · cited 2×
“…is an RNN model trained on a subset of UniProt; one key finding from the UniRep work is that using the pre-training task (in their case, next token prediction) to fine-tune on additional examples of proteins from a particular family, increases performance on down-stream tasks for that family [1, 5].”
MethodTransformer protein language models are unsupervised structure learners
Rao, R., Meier, J., Sercu, T., Ovchinnikov, S., & Rives, A. (2020) · doi:10.1101/2020.12.15.422761 · Cited by 341
An evolution-based model for designing chorismate mutase enzymes
Russ, W. P., Figliuzzi, M., Stocker, C., Barrat-Charlaix, P., Socolich, M., Kast, P., Hilvert, D., Monasson, R., Cocco, S., Weigt, M., et al. (2020) · Science · doi:10.1126/science.aba3304 · Cited by 313 · cited 10×
“[25].”
BackgroundMethodResultInfluential
Cited by
ProtGPT2 is a deep unsupervised language model for protein design
Ferruz, N., Schmidt, S., & Höcker, B. (2022) · Nature Communications · doi:10.1038/s41467-022-32007-7 · Cited by 795
Computational scoring and experimental evaluation of enzymes generated by neural networks
Johnson, S. R., Fu, X., Viknander, S., Goldin, C., Monaco, S., Zelezniak, A., & Yang, K. K. (2025) · Nature Biotechnology · doi:10.1038/s41587-024-02214-2 · Cited by 99
Self-citationGenerative power of a protein language model trained on multiple sequence alignments
(2023) · doi:10.7554/elife.79854 · Cited by 43
(2025) · doi:10.1021/jacsau.4c01101 · Cited by 14
Metadata
- Authors
- Johnson, S. R., Monaco, S., Massie, K., & Syed, Z.
- Venue
- bioRxiv (Cold Spring Harbor Laboratory)
- Publisher
- openRxiv
- Published
- 2021-01-27
- OpenAlex
- W3125063344
- Cited by
- 15
- References
- 13
- Type
- posted content
- Language
- en
- Open access
- green
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.