AI 造蛋白：比 AlphaFold3 快 10 倍，还能自己试错
蛋白质改造是场大海捞针：序列空间天文数字般大，实验只能试一小撮。ProTEUS 把每个突变当成一次「动作」，在融合了序列、进化、结构信息的空间里预测后果，再用物理约束把预测速度做到 AlphaFold3 的 10 倍以上。更关键的是闭环：它预测一批，实验测一批，结果喂回去修正模型，像 AI 自己长出了实验室。在两种酶上，它筛出的变体活性比野生型高 10 倍以上。这不是你明天能用的工具，但它是「AI 当科学家」从演示走向工程的一步。
📄 原文摘要(英文)
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.