CoChat
Sign Up
PreprintCited by 10Protein Structure and DynamicsAdvanced NMR Techniques and ApplicationsEnzyme Structure and Function

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Assisting and Accelerating NMR Assignment with Restrained Structure Prediction

Open access
Liu, Sirui, Chu, Haotian, Xie, Yuhao, Wu, Fangming, Ni, Ningxi, Wang, Chenghao, Mu, Fangjing, Wei, Jiachen, Zhang, Jun, Chen, Mengyun, Li, Junbin, Yu, Fan, Fu, Hui, Wang, Shenlin, Tian, Changlin, Wang, Zidong, Gao, Yi Qin·2023·Communications Biology
DOI 10.1101/2023.04.14.536890PMC PMC12274491

Preprint — published as 10.1038/s42003-025-08466-1 · Paper page

Open this paper in CoChat — free account

Abstract

NMR experiments can detect in situ structures and dynamic interactions, but the NMR assignment process requires expertise and is time-consuming, thereby limiting its applicability. Deep learning algorithms have been employed to aid in experimental data analysis. In this work, we developed a RASP model which can enhance structure prediction with restraints. Based on the Evoformer and structure module architecture of AlphaFold, this model can predict structure based on sequence and a flexible number of input restraints. Moreover, it can evaluate the consistency between the predicted structure and the imposed restraints. Based on this model, we constructed an iterative NMR NOESY peak assignment pipeline named FAAST, to accelerate assignment process of NOESY restraints and obtaining high quality structure ensemble. The RASP model and FAAST pipeline not only allow for the leveraging of experimental restraints to improve model prediction, but can also facilitate and expedite experimental data analysis with their integrated capabilities.

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 · 10.1101/2023.04.14.536890

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 RASP model takes in restraints in different forms and outperforms AlphaFold with restraint assistance. a A scheme of how the pairwise restraints can be taken into the model in the form of MSA, pair, IPA, and structure biases, adopted from the Evoformer and structure module in AlphaFold. b More detailed illustration of the model structure with restraints treated as biases in the Evoformer block (left) and structure block (right). For the PSP validation dataset, RASP with a fixed number of randomly sampled input restraints is compared with AlphaFold on (c) the TM-score between the PDB structure and predicted structure and d on restraint recall.
Fig. 1 RASP model takes in restraints in different forms and outperforms AlphaFold with restraint assistance. a A scheme of how the pairwise restraints can be taken into the model in the form of MSA, pair, IPA, and structure biases, adopted from the Evoformer and structure module in AlphaFold. b More detailed illustration of the model structure with restraints treated as biases in the Evoformer block (left) and structure block (right). For the PSP validation dataset, RASP with a fixed number of randomly sampled input restraints is compared with AlphaFold on (c) the TM-score between the PDB structure and predicted structure and d on restraint recall.CC BY-NC-ND · PubMed Central
Fig. 2 Performance of RASP with varied restraint numbers and quality. a Violin plot for the TM-score distributions of AlphaFold2, MEGA-Fold, and RASP with increasing number of input restraints from 0 to 200. b Violin plot for the restraint recall of AlphaFold2, MEGA-Fold, and RASP with different numbers of input restraints. The restraint recall of AF2, MEGA-Fold, and RASP with zero restraint are computed with the same restraints as from RASP with 100 restraints. c Scatter plot of the pLDDT of RASP with varied restraints number and the real TM-score of the predicted structure, with an overall correlation coefficient of 0.71. d The distribution of correlation coefficient between the RASP predicted pLDDT confidence and TM-score for the same case with different number of restraints. The distribution is drawn for all validation samples. e Violin plot for the TM-score distribution with decreasing percentage of good restraints, with total input restraints number fixed at 20. f Scatter plot of RASP pLDDT confidence and the TM-scores when loose restraints are present, with an overall correlation coefficient of 0.73. For (a, b, e), the central white dot indicates the median, the bar edges indicate the quantiles, and the whiskers extend to 1.5 times of the interquartile range. For subfigures (a, b, d, e), statistics were performed on 490 samples from the PSP validation dataset with different methods or restraints.
Fig. 2 Performance of RASP with varied restraint numbers and quality. a Violin plot for the TM-score distributions of AlphaFold2, MEGA-Fold, and RASP with increasing number of input restraints from 0 to 200. b Violin plot for the restraint recall of AlphaFold2, MEGA-Fold, and RASP with different numbers of input restraints. The restraint recall of AF2, MEGA-Fold, and RASP with zero restraint are computed with the same restraints as from RASP with 100 restraints. c Scatter plot of the pLDDT of RASP with varied restraints number and the real TM-score of the predicted structure, with an overall correlation coefficient of 0.71. d The distribution of correlation coefficient between the RASP predicted pLDDT confidence and TM-score for the same case with different number of restraints. The distribution is drawn for all validation samples. e Violin plot for the TM-score distribution with decreasing percentage of good restraints, with total input restraints number fixed at 20. f Scatter plot of RASP pLDDT confidence and the TM-scores when loose restraints are present, with an overall correlation coefficient of 0.73. For (a, b, e), the central white dot indicates the median, the bar edges indicate the quantiles, and the whiskers extend to 1.5 times of the interquartile range. For subfigures (a, b, d, e), statistics were performed on 490 samples from the PSP validation dataset with different methods or restraints.CC BY-NC-ND · PubMed Central
Fig. 3 Input restraints assist RASP prediction for multi-domain proteins, few-MSA proteins, and NMR structures. a Structures predicted by AF2 (colored blue), MEGA-Fold (colored olive) and RASP (colored salmon) for multi-domain structure 6XMV. The PDB structure of 6XMV is drawn in gray for reference. b Structures predicted by AF2 (colored blue), MEGA-Fold (colored olive) and RASP (colored salmon) for 7NBV which has few MSAs. The PDB structure of 7NBV is drawn in gray for reference. RMSDs for the three predictions are indicated. c Violin plot of overall and long-range (sequence separation≥4) restraint recall distributions and d TM-score distributions for AF2, AlphaLink in binary mode and distogram mode, and RASP. The restraint recall of deposited PDB structures are also drawn as reference. For (c, d), each box contains 333 samples, the central white dot indicates the median, the bar edges indicate the quantiles, and the whiskers extend to 1.5 times of the interquartile range.
Fig. 3 Input restraints assist RASP prediction for multi-domain proteins, few-MSA proteins, and NMR structures. a Structures predicted by AF2 (colored blue), MEGA-Fold (colored olive) and RASP (colored salmon) for multi-domain structure 6XMV. The PDB structure of 6XMV is drawn in gray for reference. b Structures predicted by AF2 (colored blue), MEGA-Fold (colored olive) and RASP (colored salmon) for 7NBV which has few MSAs. The PDB structure of 7NBV is drawn in gray for reference. RMSDs for the three predictions are indicated. c Violin plot of overall and long-range (sequence separation≥4) restraint recall distributions and d TM-score distributions for AF2, AlphaLink in binary mode and distogram mode, and RASP. The restraint recall of deposited PDB structures are also drawn as reference. For (c, d), each box contains 333 samples, the central white dot indicates the median, the bar edges indicate the quantiles, and the whiskers extend to 1.5 times of the interquartile range.CC BY-NC-ND · PubMed Central
Fig. 4 Benchmark of the FAAST NOESY assignment pipeline. a The schematic figure of the FAAST workflow and b the current assignment pipeline and parameters applied. In each iteration block, the first parameter is the percentage of input restraints subsampled for RASP structure prediction in order to construct a structure bundle, the second parameter is the cumulative probability to rule out ambiguous restraints in assignment, which is also adjustable in ARIA. c The distribution for number of NOE assignments per residue for the 57 ARTINA samples. The median number is around 15, similar with peak lists from the BMRB database. d The distribution for pairwise mutual RMSD for C-α atoms of the structure bundle. Statistics for (c, d) are performed on all 57 available samples. e The restraint recall of the FAAST structure with the identified restraints compared to the deposited PDB with the deposited NMR restraints. f Scatter plot of backbone RMSDs of FAAST structures and those reported by ARTINA.
Fig. 4 Benchmark of the FAAST NOESY assignment pipeline. a The schematic figure of the FAAST workflow and b the current assignment pipeline and parameters applied. In each iteration block, the first parameter is the percentage of input restraints subsampled for RASP structure prediction in order to construct a structure bundle, the second parameter is the cumulative probability to rule out ambiguous restraints in assignment, which is also adjustable in ARIA. c The distribution for number of NOE assignments per residue for the 57 ARTINA samples. The median number is around 15, similar with peak lists from the BMRB database. d The distribution for pairwise mutual RMSD for C-α atoms of the structure bundle. Statistics for (c, d) are performed on all 57 available samples. e The restraint recall of the FAAST structure with the identified restraints compared to the deposited PDB with the deposited NMR restraints. f Scatter plot of backbone RMSDs of FAAST structures and those reported by ARTINA.CC BY-NC-ND · 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 1 references

Research works printed but not linked: 29

  1. 2011, NMR studies of protein structure and dynamics[J], Journal of magnetic resonance, 213, 477, 10.1016/j.jmr.2011.09.009 · DOI
  2. 1990, Protein structure determination in solution by NMR spectroscopy[J], Journal of Biological Chemistry, 265, 22059, 10.1016/S0021-9258(18)45665-7 · DOI
  3. 2022, Rapid protein assignments and structures from raw NMR spectra with the deep learning technique ARTINA[J], Nature Communications, 13, 6151, 10.1038/s41467-022-33879-5 · DOI
  4. 10.1007/s10858-015-9924-9 · DOI
  5. 10.1006/jmbi.1997.1044 · DOI
  6. 10.1093/bioinformatics/btl589 · DOI
  7. 10.1016/S0022-2836(02)00241-3 · DOI
  8. 2019, Protein structure prediction using sparse NOE and RDC restraints with Rosetta in CASP13[J], Proteins, 87, 1341, 10.1002/prot.25769 · DOI
  9. 10.1016/j.str.2019.08.012 · DOI
  10. 2016, Structure prediction using sparse simulated NOE restraints with Rosetta in CASP11, Proteins, 84, 181
  11. Evans R , O’Neill M , Pritzel A , et al. Protein complex prediction with AlphaFold-Multimer[J]. BioRxiv, 2021, biorxiv: 2021.10.04.463034. · DOI · Paper page
  12. 2022, Improved AlphaFold modeling with implicit experimental information[J], Nature Methods, 19, 1376, 10.1038/s41592-022-01645-6 · DOI · Paper page
  13. Terwilliger T C , Afonine P V , Liebschner D , et al. Accelerating crystal structure determination with iterative AlphaFold prediction[J]. Acta Crystallographica Section D: Structural Biology, 2023, 79(3). · DOI
  14. Stahl K , Graziadei A , Dau T , et al. Protein structure prediction with in-cell photo-crosslinking mass spectrometry and deep learning[J]. Nature Biotechnology, 2023: 1–10. · DOI
  15. Liu S , Zhang J , Chu H , et al. PSP: million-level protein sequence dataset for protein structure prediction[J]. arXiv preprint arXiv:2206.12240, 2022. · arXiv
  16. 2021, Continuous Automated Model EvaluatiOn (CAMEO)—Perspectives on the future of fully automated evaluation of structure prediction methods[J], Proteins: Structure, Function, and Bioinformatics, 89, 1977, 10.1002/prot.26213 · DOI
  17. 2019, Introducing “best single template” models as reference baseline for the Continuous Automated Model Evaluation (CAMEO)[J], Proteins: Structure, Function, and Bioinformatics, 87, 1378, 10.1002/prot.25815 · DOI
  18. 2018, Continuous Automated Model EvaluatiOn (CAMEO) complementing the critical assessment of structure prediction in CASP12[J], Proteins: Structure, Function, and Bioinformatics, 86, 387, 10.1002/prot.25431 · DOI · Paper page
  19. 10.1093/database/bat031 · DOI
  20. 10.1002/prot.20264 · DOI · Paper page
  21. 10.1093/bioinformatics/btq066 · DOI · Paper page
  22. 2022, AlphaFold Models of Small Proteins Rival the Accuracy of Solution NMR Structures[J], Frontiers in Molecular Biosciences, 9, 877000, 10.3389/fmolb.2022.877000 · DOI · Paper page
  23. Li E H , Spaman L , Tejero R , et al. Blind Assessment of Monomeric AlphaFold2 Protein Structure Models with Experimental NMR Data[J]. BioRxiv preprint, 2023, biorxiv: 2023.01.22.525096. · DOI
  24. 2020, A method for validating the accuracy of NMR protein structures[J], Nature Communications, 11, 6321, 10.1038/s41467-020-20177-1 · DOI
  25. 10.1007/s10858-015-9955-2 · DOI
  26. 2017, Openmm 7: Rapid development of high performance algorithms for molecular dynamics[J], PLOS Computational Biology, 13, 1
  27. 10.1093/nar/28.1.235 · DOI · Paper page
  28. 2023, Biological Magnetic Resonance Data Bank[J], Nucleic Acids Research, 51, D368, 10.1093/nar/gkac1050 · DOI
  29. 10.1093/nar/gki524 · DOI · Paper page

Metadata

Authors
Liu, S., Chu, H., Xie, Y., Wu, F., Ni, N., Wang, C., Mu, F., Wei, J., Zhang, J., Chen, M., Li, J., Yu, F., Fu, H., Wang, S., Tian, C., Wang, Z., & Gao, Y. Q.
Affiliations
  • Changping Laboratory (CN)
  • Huawei Technologies (China) (CN)
  • Chinese Academy of Sciences (CN)
  • Peking University (CN)
  • East China University of Science and Technology (CN)
  • University of Science and Technology of China (CN)
Corresponding author
  • Liu, Sirui
  • Wang, Zidong
  • Gao, Yi Qin
ORCID
Venue
Communications Biology
Publisher
openRxiv
Published
2023-04-17
Volume
8
Issue
1
Pages
1067-1067
ISSN
2399-3642 (ISSN-L)
DOI
10.1101/2023.04.14.536890
PMC
PMC12274491
OpenAlex
W4412521537
Cited by
10
References
1
Citations by year
2026: 5 · 2025: 4 · 2024: 1
FWCI
2.69
Type
journal article
Language
en
Open access
Licence
cc-by-nc-nd
Locations
  • Communications Biology (publishedVersion, cc-by-nc-nd)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Natural Science Foundation of China (92053202)
  • National Science and Technology Major Project
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
  • Physical Sciences › Chemistry › Spectroscopy › Advanced NMR Techniques and Applications
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Structural Biology › Enzyme Structure and Function
Keywords
NMR resonance assignment, NOESY, protein dynamics, multidomain proteins, structure ensemble, deep learning
MeSH terms
Models, Molecular, Magnetic Resonance Spectroscopy, Protein Conformation, Proteins, Protein Folding, Nuclear Magnetic Resonance, Biomolecular

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: Assisting and Accelerating NMR Assignment with Restrained Structure Prediction. https://app.cochat.ai/papers/10.1101/2023.04.14.536890
The paper
Liu, S., Chu, H., Xie, Y., Wu, F., Ni, N., Wang, C., Mu, F., Wei, J., Zhang, J., Chen, M., Li, J., Yu, F., Fu, H., Wang, S., Tian, C., Wang, Z., & Gao, Y. Q. (2023). Assisting and Accelerating NMR Assignment with Restrained Structure Prediction. Communications Biology. https://doi.org/10.1101/2023.04.14.536890