AMix-1: A Pathway to Test-Time Scalable Protein Foundation Model
This work addresses protein engineering for biotechnology by providing a scalable model that unifies design and enables lab-in-the-loop applications, though it builds incrementally on existing methods like Bayesian Flow Networks and MSA-based strategies.
The paper tackled the challenge of developing a scalable protein foundation model for protein design, resulting in AMix-1, which generated an AmeR variant with up to 50x increased activity and demonstrated scalable performance gains through test-time algorithms.
We introduce AMix-1, a powerful protein foundation model built on Bayesian Flow Networks and empowered by a systematic training methodology, encompassing pretraining scaling laws, emergent capability analysis, in-context learning mechanism, and test-time scaling algorithm. To guarantee robust scalability, we establish a predictive scaling law and reveal the progressive emergence of structural understanding via loss perspective, culminating in a strong 1.7-billion model. Building on this foundation, we devise a multiple sequence alignment (MSA)-based in-context learning strategy to unify protein design into a general framework, where AMix-1 recognizes deep evolutionary signals among MSAs and consistently generates structurally and functionally coherent proteins. This framework enables the successful design of a dramatically improved AmeR variant with an up to $50\times$ activity increase over its wild type. Pushing the boundaries of protein engineering, we further empower AMix-1 with an evolutionary test-time scaling algorithm for in silico directed evolution that delivers substantial, scalable performance gains as verification budgets are intensified, laying the groundwork for next-generation lab-in-the-loop protein design.