Transformer protein language models are unsupervised structure learners

Roshan Rao, Joshua Meier, Tom Sercu, Sergey Ovchinnikov, Alexander Rives

Keywords: proteins, language modeling, structure prediction, unsupervised learning, explainable

Abstract: Unsupervised contact prediction is central to uncovering physical, structural, and functional constraints for protein structure determination and design. For decades, the predominant approach has been to infer evolutionary constraints from a set of related sequences. In the past year, protein language models have emerged as a potential alternative, but performance has fallen short of state-of-the-art approaches in bioinformatics. In this paper we demonstrate that Transformer attention maps learn contacts from the unsupervised language modeling objective. We find the highest capacity models that have been trained to date already outperform a state-of-the-art unsupervised contact prediction pipeline, suggesting these pipelines can be replaced with a single forward pass of an end-to-end model.

One-sentence Summary: Transformer attention maps directly represent protein contacts with state-of-the-art unsupervised precision.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics