# Topological deep learning for enhancing peptide-protein complex prediction

## TL;DR

This paper introduces TopoDockQ and ResidueX to improve peptide-protein complex predictions and design.

## Contribution

TopoDockQ uses topological deep learning to reduce false positives in peptide-protein modeling.

## Key findings

- TopoDockQ reduces false positives by 42% compared to AlphaFold2's confidence score.
- ResidueX enables integration of non-canonical amino acids into peptide scaffolds.
- The tools enhance precision and recall in peptide-protein complex predictions.

## Abstract

Peptide-protein interactions are essential to biological processes and drug discovery, but selecting high-quality models from predicted complexes remains challenging due to high false positive rates (FPR). Here we introduce TopoDockQ, a topological deep learning model leveraging persistent combinatorial Laplacian (PCL) features to predict DockQ scores (p-DockQ) for accurately evaluating peptide-protein interface quality, aimed at enhancing precision and mitigating FPR in model selection. Compared to AlphaFold2’s built-in confidence score, TopoDockQ reduces false positives by at least 42% and increases precision by 6.7% across five evaluation datasets filtered to ≤70% peptide-protein sequence identity, while maintaining relatively high recall and F1 scores. To support flexible peptide design, we introduce ResidueX, a workflow incorporating non-canonical amino acids (ncAA) into peptide scaffolds. Together, TopoDockQ and ResidueX advance peptide-protein modeling by refining confidence scoring and supporting ncAA incorporation, enabling precise, customizable design and accelerating next-generation peptide therapeutics development.

Peptide-protein interactions are crucial for biological processes, yet accurate structural modeling remains challenging. Here, the authors introduce TopoDockQ, a topological deep learning model to enhance model selection, and ResidueX, a workflow for non-canonical amino acids integrating into custom peptide scaffolds, which represent a synergistic advancement in peptide-protein modeling and enhance more precise and versatile peptide design.