q-bio.BMNov 9, 2026

ProTEUS: A mutation-resolved world model for closed-loop protein evolution

Authors: Yikun Zhang, Minghao Guo, Guangxue Xu, Peter Yichen Chen, Wojciech Matusik, Wengong Jin, Jingjing Chai, Kaiyi Jiang

Organizations: Foundry Bioscience Inc. Burlingame, CA, 94010, United States of America · Khoury College of Computer Sciences Northeastern University, Boston, MA, 02115, United States of America · Broad Institute of MIT and Harvard Cambridge, MA, 02142, United States of America · Department of Computer Science Massachusetts Institute of Technology Cambridge, MA, 02139, United States of America · 

Abstract

Protein engineering requires navigating rugged fitness landscapes, where small sequence changes can produce large functional differences, but experimental screening can only sample a small fraction of sequence space. Thus, a central challenge is to efficiently traverse the large combinatorial space of mutations in silico. We introduce ProTEUS, a mutation-resolved world model for closed-loop protein evolution. Pretrained on deep mutational scanning assays, ProTEUS learns a joint representation of sequence, evolutionary, and structural information using MutationFormer, a mutation-biased attention architecture. Variants are represented as states in this joint latent space, and mutations define actions whose functional consequences are predicted. To support exploration of many mutants, we developed a fast, physics-constrained all-atom predictor that utilizes explicit geometric constraint enforcement to achieve >10 fold increase in speed than AlphaFold3 with guaranteed physical validity. Lastly, ProTEUS refines its prediction and understanding of protein activity through closed loop experimental feedback to guide evolution toward a desired phenotype. ProTEUS surpasses established models in both zero-shot and supervised fine-tuning mutation prediction tasks. We experimentally demonstrate its utility in two closed-loop multi-objective engineering campaigns on aldehyde dehydrogenase and levansucrase, yielding variants with >10 fold higher activity than wild type. ProTEUS provides a world-model framework that connects virtual evaluation of mutational outcomes with experimental learning for efficient protein evolution.

Explore similar work

CardsList
  1. MutAtlas: A PDB-Wide Energy-Guided Atlas of Protein Mutation Effects

    Nov 12, 2026Ruihan Guo, Chaoran Cheng, Zhanghan Ni +3protein mutation effect predictionProtein Fitness Prediction

  2. EvoFlows: Evolutionary Edit-Based Flow-Matching for Protein Engineering

    Mar 12, 2026Nicolas Deutschmann, Constance Ferragu, Jonathan D. Ziegler +2Protein DesignSequence Modeling

  3. Better Protein Function Prediction by Modeling Survivorship Bias

    May 7, 2026Zhongmou Chao, Poompol Buathong, Ekaterina Selivanovitch +2Protein Function PredictionWeakly Supervised Learning