Identification and engineering of highly functional potyviral proteases in cells using co-evolutionary models

2026-02-26 · Source: Machine learning : nature.com subject feeds · Field: Science & Research — Life Sciences & Biology, Mathematics & Computational Sciences, Engineering & Applied Sciences · Depth: Expert, medium

Summary

Researchers have developed a co-evolutionary model to predict and engineer highly functional proteases from the Potyviridae family, addressing the previously unprofiled efficiency and substrate specificity of these enzymes. The model accurately predicts protease performance at single amino-acid resolution, leading to the identification and engineering of several proteases that outperform the commercially available tobacco etch virus protease (TEVp). To showcase the method's precision, the team engineered protease crosstalk to selectively activate a synthetic cell-death program in human cells. The study's data, including aligned Potyviridae sequences and plasmid lists, are available on Zenodo (accession code: https://doi.org/10.5281/zenodo.15039890) and Addgene. An interactive web application, ProSSpeC (https://coevolutionary.org/prosspec/), and its code (https://github.com/morcoslab/ProSSpeC) have been released to facilitate further testing of potyviral proteases.

Key takeaway

For AI Scientists developing synthetic biology tools, this research demonstrates a robust method for engineering highly specific proteases. Your work on designing complex cellular circuits could benefit from integrating co-evolutionary modeling to predict and optimize enzyme function, potentially enabling more precise control over biological processes like targeted cell death. Consider exploring the ProSSpeC web application to accelerate your protease design and validation efforts.

Key insights

Co-evolutionary models can accurately predict and engineer superior potyviral proteases for precise cellular control.

Principles

• Co-evolutionary features predict protease performance.
• Engineered proteases can exceed commercial standards.

Method

A co-evolutionary model learns features to predict protease performance at single amino-acid resolution, enabling the identification and engineering of proteases with enhanced efficiency and specificity.

In practice

• Utilize ProSSpeC web app for protease testing.
• Access plasmid sequences on Addgene for research.

Topics

• Potyviral Proteases
• Co-evolutionary Models
• Protease Engineering
• Synthetic Biology
• ProSSpeC Software

Code references

• morcoslab/ProSSpeC

Best for: AI Scientist, AI Researcher, Research Scientist

Related on AIssential

• A geometric foundation model for enzyme retrieval with evolutionary insights — EnzymeCAGE is a geometric foundation model that predicts enzyme function and reaction specificity using structural and evolutionary data.
• Profluent and Lilly: the next gene editor will be designed by AI — AI-designed recombinases offer a credible path to kilobase-scale DNA editing for complex genetic diseases.
• A Self-Evolving Agentic System for Automated Generation and Execution of Biological Protocols — ProtoPilot demonstrates a verifiable route to autonomous wet-lab experimentation through a self-evolving multi-agent system.
• Single-round evolution of RNA aptamers with GRAPE-LM — GRAPE-LM enables single-round RNA aptamer evolution using AI, outperforming multi-round human methods.
• GPT-5 lowers the cost of cell-free protein synthesis — GPT-5 connected to a robotic lab significantly reduced cell-free protein synthesis costs through autonomous closed-loop optimization.
• Researchers try to cut the genetic code from 20 to 19 amino acids — Researchers from Columbia and Harvard successfully engineered *E.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.