Computational design of target-specific linear peptide binders with TransformerBeta

TL;DR

This paper introduces TransformerBeta, a novel machine learning approach using Transformer architecture to computationally design target-specific linear peptide binders based on large-scale structural data, aiding protein engineering.

Contribution

The study develops TransformerBeta, a new method leveraging AlphaFold structures and Transformer models for designing specific peptide binders, addressing data scarcity and dynamic nature challenges.

Findings

TransformerBeta accurately predicts beta strand interactions.

It samples sequences with beta sheet-like properties.

The method captures interpretable physico-chemical interaction patterns.

Abstract

The computational prediction and design of peptide binders targeting specific linear epitopes is crucial in biological and biomedical research, yet it remains challenging due to their highly dynamic nature and the scarcity of experimentally solved binding data. To address this problem, we built an unprecedentedly large-scale library of peptide pairs within stable secondary structures (beta sheets), leveraging newly available AlphaFold predicted structures. We then developed a machine learning method based on the Transformer architecture for the design of specific linear binders, in analogy to a language translation task. Our method, TransformerBeta, accurately predicts specific beta strand interactions and samples sequences with beta sheet-like molecular properties, while capturing interpretable physico-chemical interaction patterns. As such, it can propose specific candidate binders targeting linear epitope for experimental validation to inform protein design.