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.

  • Research Briefing
  • Published:

Assessing the laboratory performance of AI-generated enzymes

A set of 20 computational metrics was evaluated to determine whether they could predict the functionality of synthetic enzyme sequences produced by generative protein models, resulting in the development of a computational filter, COMPSS, that increased experimental success rates by 50–150%, tested in over 500 natural and AI-generated enzymes.

This is a preview of subscription content, access via your institution

Access options

Buy this article

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Benchmarking in silico metrics for prediction of enzyme functionality.


  1. Repecka, D. et al. Expanding functional protein sequence spaces using generative adversarial networks. Nat. Mach. Intell. 3, 324–333 (2021). Among the first experimentally validated generative models of protein sequences demonstrating that AI can generate diverse functional enzymes.

  2. Meier, J. et al. Language models enable zero-shot prediction of the effects of mutations on protein function. Preprint at bioRxiv (2021). The paper presents one of the top-performing models that ended up in the COMPSS filter.

  3. Dauparas, J. et al. Robust deep learning-based protein sequence design using ProteinMPNN. Science 378, 49–56 (2022). The paper presents one of the top-performing models that ended up in the COMPSS filter.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Madani, A. et al. Large language models generate functional protein sequences across diverse families. Nat. Biotechnol. 41, 1099–1106 (2023). A recent generative sequence model example that is based on a large protein language transformer.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Ingraham, J. B. et al. Illuminating protein space with a programmable generative model. Nature 623, 1070–1078 (2023). A paper showing the successful application of generative diffusion models conditioned on geometrical protein properties.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This is a summary of: Johnson, S. R. et al. Computational scoring and experimental evaluation of enzymes generated by neural networks. Nat. Biotechnol. (2024).

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Assessing the laboratory performance of AI-generated enzymes. Nat Biotechnol (2024).

Download citation

  • Published:

  • DOI:


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