Scholarlink

In recent years, deep learning techniques have enabled the development of numerous methods for protein contact prediction, achieving significant success. This study introduces a protein contact prediction model based on a deep residual network, named Se-Inter. Unlike previous deep learning approaches, we leverage structure-informed residue sequences combined with protein language models to construct features. We conducted extensive evaluations of Se-Inter on multiple test datasets and compared it with several mainstream protein contact prediction methods, including DeepHomo, GLINTER, and DRN-1D2D-Inter. The results demonstrate the accuracy and robustness of Se-Inter.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Protein-Protein Contact Prediction Using Structure-Informed Sequences and Protein Language Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Qiyuan Huang,
  • Lei Wang,
  • Zhihong Zhang,
  • Jingjing Zhao,
  • Zhidong Xue

摘要

In recent years, deep learning techniques have enabled the development of numerous methods for protein contact prediction, achieving significant success. This study introduces a protein contact prediction model based on a deep residual network, named Se-Inter. Unlike previous deep learning approaches, we leverage structure-informed residue sequences combined with protein language models to construct features. We conducted extensive evaluations of Se-Inter on multiple test datasets and compared it with several mainstream protein contact prediction methods, including DeepHomo, GLINTER, and DRN-1D2D-Inter. The results demonstrate the accuracy and robustness of Se-Inter.