The generative capacity of probabilistic protein sequence models
Open accessThe DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index.
Abstract
Potts models and variational autoencoders (VAEs) have recently gained popularity as generative protein sequence models (GPSMs) to explore fitness landscapes and predict mutation effects. Despite encouraging results, current model evaluation metrics leave unclear whether GPSMs faithfully reproduce the complex multi-residue mutational patterns observed in natural sequences due to epistasis. Here, we develop a set of sequence statistics to assess the “generative capacity” of three current GPSMs: the pairwise Potts Hamiltonian, the VAE, and the site-independent model. We show that the Potts model’s generative capacity is largest, as the higher-order mutational statistics generated by the model agree with those observed for natural sequences, while the VAE’s lies between the Potts and site-independent models. Importantly, our work provides a new framework for evaluating and interpreting GPSM accuracy which emphasizes the role of higher-order covariation and epistasis, with broader implications for probabilistic sequence models in general.
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
Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., Wieser, E., Taylor, J., Berg, S., Smith, N. J., et al. (2020) · Nature · doi:10.1038/s41586-020-2649-2 · Cited by 23,826
Self-citationUniProt: a worldwide hub of protein knowledge
The UniProt Consortium (2019) · Nucleic Acids Research · doi:10.1093/nar/gky1049 · Cited by 7,698 · cited 3×
“Sequence dataset preparation For the natural analysis, we use an MSA of the kinase protein superfamily which we have previously curated using sequences from the Uniprot/TREMBL database.(40) This MSA is composed of ∼20K sequences of length 232, obtained by filtering a larger set of ∼291K sequences to remove any sequenc…”
MethodInfluentialThe Pfam protein families database in 2019
El-Gebali, S., Mistry, J., Bateman, A., Eddy, S. R., Luciani, A., Potter, S. C., Qureshi, M., Richardson, L. J., Salazar, G. A., Smart, A., et al. (2019) · Nucleic Acids Research · doi:10.1093/nar/gky995 · Cited by 4,602 · cited 4×
“…thus how well the GPSMs generalize, using two training MSA sizes: one representing the estimated size upper bound for a Pfam protein family (10K) (see Supplementary Information), 38 and one being large enough to eliminate out-of-sample error in our generative capacity measurements (see Results).”
BackgroundMethodInfluentialRives, 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
An Introduction to Variational Autoencoders
Kingma, D. P., & Welling, M. (2019) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000056 · Cited by 2,609
HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment
Remmert, M., Biegert, A., Hauser, A., & Söding, J. (2012) · Nature Methods · doi:10.1038/nmeth.1818 · Cited by 2,288
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
Foundations of Machine Learning
Mohri, M., Rostamizadeh, A., & Talwalkar, A. (2018) · doi:10.7551/mitpress/10290.001.0001 · Cited by 1,122
Mutation effects predicted from sequence co-variation
Hopf, T. A., Ingraham, J. B., Poelwijk, F. J., Schärfe, C. P. I., Springer, M., Sander, C., & Marks, D. S. (2017) · Nature Biotechnology · doi:10.1038/nbt.3769 · Cited by 775 · cited 1×
“11,12 They are “generative” in the sense that they define the probability, p ( S ) , that a protein sequence S results from the evolutionary process.”
Background
Cited by
From sequence to function through structure: Deep learning for protein design
(2022) · doi:10.1016/j.csbj.2022.11.014 · Cited by 124
Protein language models trained on multiple sequence alignments learn phylogenetic relationships
Lupo, U., Sgarbossa, D., & Bitbol, A.-F. (2022) · Nature Communications · doi:10.1038/s41467-022-34032-y · Cited by 67
Generative power of a protein language model trained on multiple sequence alignments
(2023) · doi:10.7554/elife.79854 · Cited by 43
Malbranke, C., Rostain, W., Depardieu, F., Cocco, S., Monasson, R., & Bikard, D. (2023) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1011621 · Cited by 15
(2025) · doi:10.1021/jacsau.4c01101 · Cited by 14
Metadata
- Authors
- McGee, F., Hauri, S., Novinger, Q., Vucetic, S., Levy, R. M., Carnevale, V., & Haldane, A.
- Venue
- Nature Communications
- Published
- 2021-11-02
- Volume
- 12
- Issue
- 1
- Pages
- 6302-6302
- ISSN
- 2041-1723 (ISSN-L)
- OpenAlex
- W3208082951
- Cited by
- 54
- References
- 23
- FWCI
- 5.04
- Type
- journal article
- Language
- en
- Open access
- gold
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: The generative capacity of probabilistic protein sequence models. https://app.cochat.ai/papers/10.1038/s41467-021-26529-9
- The paper
- McGee, F., Hauri, S., Novinger, Q., Vucetic, S., Levy, R. M., Carnevale, V., & Haldane, A. (2021). The generative capacity of probabilistic protein sequence models. Nature Communications. https://doi.org/10.1038/s41467-021-26529-9
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.