CoChat
Sign Up
Cited by 447Protein Structure and DynamicsMachine Learning in BioinformaticsComputational Drug Discovery Methods

The DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.

Illuminating protein space with a programmable generative model

Open access
Ingraham, John B., Baranov, Max, Costello, Zak, Barber, Karl W., Wang, Wujie, Ismail, Ahmed, Frappier, Vincent, Lord, Dana M., Ng-Thow-Hing, Christopher, Van Vlack, Erik R., Tie, Shan, Xue, Vincent, Cowles, Sarah C., Leung, Alan, Rodrigues, João V., Morales-Perez, Claudio L., Ayoub, Alex M., Green, Robin, Puentes, Katherine, Oplinger, Frank, Panwar, Nishant V., Obermeyer, Fritz, Root, Adam R., Beam, Andrew L., Poelwijk, Frank J., Grigoryan, Gevorg·2023·Nature
DOI 10.1038/s41586-023-06728-8PMC PMC10686827

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Three billion years of evolution has produced a tremendous diversity of protein molecules 1 , but the full potential of proteins is likely to be much greater. Accessing this potential has been challenging for both computation and experiments because the space of possible protein molecules is much larger than the space of those likely to have functions. Here we introduce Chroma, a generative model for proteins and protein complexes that can directly sample novel protein structures and sequences, and that can be conditioned to steer the generative process towards desired properties and functions. To enable this, we introduce a diffusion process that respects the conformational statistics of polymer ensembles, an efficient neural architecture for molecular systems that enables long-range reasoning with sub-quadratic scaling, layers for efficiently synthesizing three-dimensional structures of proteins from predicted inter-residue geometries and a general low-temperature sampling algorithm for diffusion models. Chroma achieves protein design as Bayesian inference under external constraints, which can involve symmetries, substructure, shape, semantics and even natural-language prompts. The experimental characterization of 310 proteins shows that sampling from Chroma results in proteins that are highly expressed, fold and have favourable biophysical properties. The crystal structures of two designed proteins exhibit atomistic agreement with Chroma samples (a backbone root-mean-square deviation of around 1.0 Å). With this unified approach to protein design, we hope to accelerate the programming of protein matter to benefit human health, materials science and synthetic biology.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41586-023-06728-8

Cites

54 of 54 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Chroma is a generative model for proteins and protein complexes that combines structured diffusion for protein backbones with scalable molecular neural networks for backbone synthesis and all-atom design. a, A correlated diffusion process with chain and radius-of-gyration constraints gradually transforms protein structures into random collapsed polymers (right to left). The reverse process (left to right) can be expressed in terms of a time-dependent optimal denoiser x^θ(xt,t) that maps noisy coordinates xt at time t to predicted denoised coordinates x0. b, We parameterize this in terms of a random graph neural network with long-range connectivity inspired by efficient N-body algorithms (middle) and a fast method for solving for a global consensus structure given predicted inter-residue geometries (right). Another graph-based design network (a, top right) generates protein sequences and side-chain conformations conditionally based on the sampled backbone. c, The time-dependent protein prior learnt by the diffusion model can be combined with composable restraints and constraints for the programmable generation of protein systems.
Fig. 1 Chroma is a generative model for proteins and protein complexes that combines structured diffusion for protein backbones with scalable molecular neural networks for backbone synthesis and all-atom design. a, A correlated diffusion process with chain and radius-of-gyration constraints gradually transforms protein structures into random collapsed polymers (right to left). The reverse process (left to right) can be expressed in terms of a time-dependent optimal denoiser x^θ(xt,t) that maps noisy coordinates xt at time t to predicted denoised coordinates x0. b, We parameterize this in terms of a random graph neural network with long-range connectivity inspired by efficient N-body algorithms (middle) and a fast method for solving for a global consensus structure given predicted inter-residue geometries (right). Another graph-based design network (a, top right) generates protein sequences and side-chain conformations conditionally based on the sampled backbone. c, The time-dependent protein prior learnt by the diffusion model can be combined with composable restraints and constraints for the programmable generation of protein systems.CC BY · PubMed Central
Fig. 2 Analysis of unconditional samples reveals diverse geometries that exhibit new higher-order structures and refold in silico. a, A representative set of Chroma-sampled proteins and protein complexes exhibits complex and diverse topologies with high secondary-structure content, including familiar TIM (triose-phosphate isomerase) barrel-like folds (top left), antibody–antigen-like complexes (centre right) and new arrangements of helical bundles and β-sheets. b,c, Despite these qualitative similarities, samples frequently have low nearest-neighbour similarity to structures in the PDB, as measured by nearest-neighbour TM score 41 (b; Supplementary Appendix J.4), with structures demonstrating frequent novelty across length ranges (c). d,e, When we attempted to refold samples in silico using only a single sequence sample per structure, we observed widespread refolding with a high degree of superposition (d), including occasionally in the very high length range of more than 800 residues (e).
Fig. 2 Analysis of unconditional samples reveals diverse geometries that exhibit new higher-order structures and refold in silico. a, A representative set of Chroma-sampled proteins and protein complexes exhibits complex and diverse topologies with high secondary-structure content, including familiar TIM (triose-phosphate isomerase) barrel-like folds (top left), antibody–antigen-like complexes (centre right) and new arrangements of helical bundles and β-sheets. b,c, Despite these qualitative similarities, samples frequently have low nearest-neighbour similarity to structures in the PDB, as measured by nearest-neighbour TM score 41 (b; Supplementary Appendix J.4), with structures demonstrating frequent novelty across length ranges (c). d,e, When we attempted to refold samples in silico using only a single sequence sample per structure, we observed widespread refolding with a high degree of superposition (d), including occasionally in the very high length range of more than 800 residues (e).CC BY · PubMed Central
Fig. 3 Symmetry, substructure and shape conditioning enable geometric molecular programming. a, Sampling oligomeric structures with arbitrary chain symmetries is possible by using a conditioner that tessellates an asymmetric subunit in the energy function. Cyclic (Cn), dihedral (Dn), tetrahedral (T), octahedral (O) and icosahedral (I) symmetry groups can produce a wide variety of possible homomeric complexes. The right-most protein complex contains 60 subunits and 60,000 total residues, which is enabled by leveraging symmetries and using our subquadratically scaling architecture. b, Conditioning on partial substructure (monochrome) enables protein infilling or outfilling. The top two rows illustrate regeneration (colour) of half a protein (the enzyme DHFR, first row) or complementarity-determining region loops of a VHH antibody (second row). The next three rows show conditioning on a predefined motif. The order and matching location of motif segments is not prespecified here. c, Conditioning on arbitrary volumetric shapes is exemplified by the complex geometries of the Latin alphabet and Arabic numerals. All structures were selected from protocols with high rates of in silico refolding (Supplementary Appendix K).
Fig. 3 Symmetry, substructure and shape conditioning enable geometric molecular programming. a, Sampling oligomeric structures with arbitrary chain symmetries is possible by using a conditioner that tessellates an asymmetric subunit in the energy function. Cyclic (Cn), dihedral (Dn), tetrahedral (T), octahedral (O) and icosahedral (I) symmetry groups can produce a wide variety of possible homomeric complexes. The right-most protein complex contains 60 subunits and 60,000 total residues, which is enabled by leveraging symmetries and using our subquadratically scaling architecture. b, Conditioning on partial substructure (monochrome) enables protein infilling or outfilling. The top two rows illustrate regeneration (colour) of half a protein (the enzyme DHFR, first row) or complementarity-determining region loops of a VHH antibody (second row). The next three rows show conditioning on a predefined motif. The order and matching location of motif segments is not prespecified here. c, Conditioning on arbitrary volumetric shapes is exemplified by the complex geometries of the Latin alphabet and Arabic numerals. All structures were selected from protocols with high rates of in silico refolding (Supplementary Appendix K).CC BY · PubMed Central
Fig. 4 Protein structure classifiers and caption models can bias the sampling process towards user-specified properties. a, Neural networks trained to predict protein properties can bias unconditional samples (top) towards states that optimize predicted properties, such as secondary-structure composition (bottom) indicated by CATH class level codes (C1, Mainly Alpha; C2, Mainly Beta; C3, Alpha Beta). b, A neural network trained to predict CATH topology annotations can routinely drive generation towards samples with high predicted probabilities of the intended class label, which sometimes aligns with our intended fold topology for highly abundant labels. Left, highly abundant Rossmann fold (CATH topology 3.40.50, 14.0% of training set); middle, highly abundant Ig fold (CATH topology 2.60.40, 9.8% of training set); right, a rare specific β-barrel fold (CATH topology 2.40.155, 0.07% of training set). c, Fine-tuning a multi-label predictor to bias a pretrained large language model into a structure caption predictor can enable natural language conditioning. We begin to see examples of semantic alignment between prompts and output structures for highly abundant classes of structures, although we do not always see this reflected in the time-zero caption perplexity (CP, lower is better). Left, ‘crystal structure of a Rossmann fold’; right, ‘crystal structure of a Fab antibody fragment’.
Fig. 4 Protein structure classifiers and caption models can bias the sampling process towards user-specified properties. a, Neural networks trained to predict protein properties can bias unconditional samples (top) towards states that optimize predicted properties, such as secondary-structure composition (bottom) indicated by CATH class level codes (C1, Mainly Alpha; C2, Mainly Beta; C3, Alpha Beta). b, A neural network trained to predict CATH topology annotations can routinely drive generation towards samples with high predicted probabilities of the intended class label, which sometimes aligns with our intended fold topology for highly abundant labels. Left, highly abundant Rossmann fold (CATH topology 3.40.50, 14.0% of training set); middle, highly abundant Ig fold (CATH topology 2.60.40, 9.8% of training set); right, a rare specific β-barrel fold (CATH topology 2.40.155, 0.07% of training set). c, Fine-tuning a multi-label predictor to bias a pretrained large language model into a structure caption predictor can enable natural language conditioning. We begin to see examples of semantic alignment between prompts and output structures for highly abundant classes of structures, although we do not always see this reflected in the time-zero caption perplexity (CP, lower is better). Left, ‘crystal structure of a Rossmann fold’; right, ‘crystal structure of a Fab antibody fragment’.CC BY · PubMed Central
Fig. 5 Experimental validation of Chroma-designed proteins. a, Protocol for protein design and experimental validation. Unconditional designs: 268 proteins. Semantic conditioning: 12 α-conditioned, 13 β-conditioned, 11 α/β mixtures and 6 with β-barrel topology. See text for details. b, Rank-ordered unconditional Chroma protein solubility scores by the split-GFP assay for 172 tested proteins. Red dots and error bars denote means and standard deviations, respectively, from three biological replicates. c,d, X-ray crystal structures (rainbow) of UNC_079 (c, 1.1 Å resolution, PDB 8TNM, root-mean-square deviation (RMSD) = 1.1 Å) and UNC_239 (d, 2.4 Å resolution, PDB 8TNO, RMSD = 1.0 Å) overlaid with Chroma-generated models (grey). Insets compare each crystal structure (rainbow) with its nearest PDB match (4NH2 and 6AFV, respectively; grey). e, CD data for seven purified Chroma proteins. The fraction of α-helical and β-strand content was determined using BeStSel50. Tm is the melting temperature determined by differential scanning calorimetry and SS designates secondary structure. f, CD data for three purified Chroma conditional designs: SEM_018 (α-conditioned), SEM_038 (β-barrel topology) and SEM_011 (α/β mixture). g,h, Correlation between predicted secondary-structure content in Chroma designs compared with the prediction from CD, for α-helical (g) and β-strand (h) content.
Fig. 5 Experimental validation of Chroma-designed proteins. a, Protocol for protein design and experimental validation. Unconditional designs: 268 proteins. Semantic conditioning: 12 α-conditioned, 13 β-conditioned, 11 α/β mixtures and 6 with β-barrel topology. See text for details. b, Rank-ordered unconditional Chroma protein solubility scores by the split-GFP assay for 172 tested proteins. Red dots and error bars denote means and standard deviations, respectively, from three biological replicates. c,d, X-ray crystal structures (rainbow) of UNC_079 (c, 1.1 Å resolution, PDB 8TNM, root-mean-square deviation (RMSD) = 1.1 Å) and UNC_239 (d, 2.4 Å resolution, PDB 8TNO, RMSD = 1.0 Å) overlaid with Chroma-generated models (grey). Insets compare each crystal structure (rainbow) with its nearest PDB match (4NH2 and 6AFV, respectively; grey). e, CD data for seven purified Chroma proteins. The fraction of α-helical and β-strand content was determined using BeStSel50. Tm is the melting temperature determined by differential scanning calorimetry and SS designates secondary structure. f, CD data for three purified Chroma conditional designs: SEM_018 (α-conditioned), SEM_038 (β-barrel topology) and SEM_011 (α/β mixture). g,h, Correlation between predicted secondary-structure content in Chroma designs compared with the prediction from CD, for α-helical (g) and β-strand (h) content.CC BY · PubMed Central

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.

A few free checks a day · papers already verified open instantly · results are public

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

Show all 45 references

Research works printed but not linked: 6

  1. Ramesh, A., Dhariwal, P., Nichol, A., Chu, C. & Chen, M. Hierarchical text-conditional image generation with CLIP latents. Preprint at https://arxiv.org/abs/2204.06125 (2022). · arXiv
  2. Ingraham, J., Garg, V., Barzilay, R. & Jaakkola, T. Generative models for graph-based protein design. In Proc. Advances in Neural Information Processing Systems 32 (eds Wallach, H. et al.) (NeurIPS, 2019).
  3. Anand, N. & Huang, P.-S. Generative modeling for protein structures. In Proc. Advances in Neural Information Processing Systems 31 (eds Bengio, S. et al.) (NeurIPS, 2018).
  4. Lin, Z., Sercu, T., LeCun, Y. & Rives, A. Deep generative models create new and diverse protein structures. In Machine Learning in Structural Biology Workshop at the 35th Conference on Neural Information Processing Systems (MLSB, 2021).
  5. Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O. & Dahl, G. E. Neural message passing for quantum chemistry. In Proc. 34th International Conference on Machine Learning (eds Precup, D. et al.) 1263–1272 (PMLR, 2017).
  6. Jing, B., Eismann, S., Suriana, P., Townshend, R. J. L. & Dror, R. Learning from protein structure with geometric vector perceptrons. In International Conference on Learning Representations (eds Hofmann, K. et al.) (OpenReview.net, 2021).

Cited by

Sorted by citations

Items 101–200 of 447

Related

See all

Metadata

Authors
Ingraham, J. B., Baranov, M., Costello, Z., Barber, K. W., Wang, W., Ismail, A., Frappier, V., Lord, D. M., Ng-Thow-Hing, C., Van Vlack, E. R., Tie, S., Xue, V., Cowles, S. C., Leung, A., Rodrigues, J. V., Morales-Perez, C. L., Ayoub, A. M., Green, R., Puentes, K., Oplinger, F., Panwar, N. V., Obermeyer, F., Root, A. R., Beam, A. L., Poelwijk, F. J., & Grigoryan, G.
Corresponding author
Grigoryan, Gevorg
ORCID
Venue
Nature
Publisher
Springer Science and Business Media LLC
Published
2023-11-15
Volume
623
Issue
7989
Pages
1070-1078
ISSN
  • 0028-0836 (ISSN-L)
  • 1476-4687
DOI
10.1038/s41586-023-06728-8
PMC
PMC10686827
OpenAlex
W4388694364
Cited by
444
References
45
Citations by year
2026: 152 · 2025: 147 · 2024: 125 · 2023: 20
FWCI
39.56
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
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 › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
Keywords
protein design, generative models, diffusion models, protein complexes, protein structure prediction, Bayesian inference, protein folding, natural language prompts, color intensity
MeSH terms
Machine Learning, Algorithms, Bayes Theorem, Computer Simulation, Humans, Models, Molecular, Protein Conformation, Proteins, Semantics, Protein Folding, Directed Molecular Evolution, Synthetic Biology

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