Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.


Rapid protein model refinement by deep learning

A graph-neural-network-based framework is proposed for the refinement of protein structure models, substantially improving the efficacy and efficiency of refining protein models when compared with the state-of-the-art approaches.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: Comparison of the GNN approach to alternative deep learning approaches.


  1. 1.

    Dill, K. A. & MacCallum, J. L. Science 338, 1042–1046 (2012).

    Article  Google Scholar 

  2. 2.

    Jumper J. et al. Critical Assessment of Techniques for Protein Structure Prediction (CASP, 2020).

  3. 3.

    Jing, X. & Xu, J. Nat. Comput. Sci. (2021).

  4. 4.

    Heo, L., Arbour, C. F. & Feig, M. Proteins 87, 1263–1275 (2019).

    Article  Google Scholar 

  5. 5.

    Park, H. et al. Proteins 87, 1276–1282 (2019).

    Article  Google Scholar 

  6. 6.

    Conway, P., Tyka, M. D., DiMaio, F., Konerding, D. E. & Baker, D. Protein Sci. 23, 47–55 (2014).

    Article  Google Scholar 

  7. 7.

    Zamora-Resendiz, R. & Crivelli, S. Preprint at bioRxiv (2019).

  8. 8.

    AlQuraishi, M. Cell Syst. 8, 292–301 (2019).

    Article  Google Scholar 

  9. 9.

    Ingraham, J. & Riesselman, A. J. In Int. Conf. Learning Representations (ICLR, 2019).

  10. 10.

    Hiranuma, N. et al. Nat. Commun. 12, 1340 (2021).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Philip M. Kim.

Ethics declarations

Competing interests

P.M.K. is a co-founder and has been consultant to several biotechnology ventures, including Resolute Bio, and serves on the scientific advisory board of ProteinQure. He also holds several patents in the area of protein and peptide engineering. O.A. declares no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Abdin, O., Kim, P.M. Rapid protein model refinement by deep learning. Nat Comput Sci 1, 456–457 (2021).

Download citation


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing