Uni-Fold MuSSe: De Novo Protein Complex Prediction with Protein Language Models

• DOI: 10.1101/2023.02.14.528571
• Publication Date: 2023-02-15T17:21:51Z
• Authors: Jinhua Zhu; Zhenyu He; Ziyao Li; Guolin Ke; Linfeng Zhang
• Keywords: Protein sequencing; Sequence (biology)

A bstract Accurately solving the structures of protein complexes is crucial for understanding and further modifying biological activities. Recent success AlphaFold its variants shows that deep learning models are capable accurately predicting complex structures, yet with painstaking effort homology search pairing. To bypass this need, we present Uni-Fold MuSSe (Multimer Single Sequence inputs), which predicts from their primary sequences aid pre-trained language models. Specifically, built prediction based on sequence representations ESM-2, a large model 3 billion parameters. In order to adapt inter-protein evolutionary patterns, slightly modified groups known interactions. Our results highlight potential suggest room improvements.