Deep Learning for Protein Complex Prediction and Design

TL;DR

This paper explores deep learning methods for predicting and designing protein complex structures, focusing on hierarchical architectures and efficient search algorithms to improve accuracy and functionality.

Contribution

It introduces novel deep learning architectures and search algorithms tailored for protein complex modeling and design, advancing computational structural biology.

Findings

Developed domain-specific deep learning architectures for protein complexes

Created search algorithms for navigating protein sequence space

Enhanced prediction accuracy and design capabilities for protein complexes

Abstract

Accurately modeling and designing protein complex structures is a central problem in computational structural biology, with broad implications for understanding cellular function and developing therapeutics. This thesis investigates two fundamental aspects of this problem using deep learning: domain-specific architectures that capture the hierarchical nature of protein structures, and search algorithms that efficiently navigate the vast sequence spaces of protein complexes to identify interacting homologs for improving complex structure prediction and to design protein sequences.