使用对称扩展变换来预测对称蛋白质结构
By iteratively refining protein structure prediction through symmetric expansion blocks and folded neural networks, the problem of insufficient utilization of protein symmetry features in existing models is solved, the prediction accuracy is improved, and biochemical research and drug development are promoted.
CN116325000BActive Publication Date: 2026-07-17GDM HOLDINGS LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GDM HOLDINGS LTD
- Filing Date
- 2021-11-23
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing machine learning models struggle to effectively utilize the symmetry features of proteins when predicting their structures, resulting in insufficient accuracy in structure prediction.
Method used
By employing symmetric expansion blocks and folded neural networks, protein structure prediction is iteratively refined, its symmetry is explicitly enforced, and the structural parameters of the amino acid chain are updated using symmetric expansion transformation, thereby improving prediction accuracy.
Benefits of technology
It significantly improves the accuracy of protein structure prediction, reduces computational resource consumption, and promotes biochemical research and drug development.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN116325000B_ABST
Abstract
用于预测包括多条氨基酸链的蛋白质的结构的方法、系统和装置，包括在计算机存储介质上编码的计算机程序。根据一个方面，一种方法包括：获得蛋白质中的第一氨基酸链的初始结构参数；获得识别对称基团的数据；使用包括更新块序列的折叠神经网络处理第一氨基酸链的初始结构参数和识别对称基团的数据，其中每个更新块执行包括以下的操作：将对称扩展变换应用于第一氨基酸链的当前结构参数；以及根据更新块的更新块参数的值来处理蛋白质中的氨基酸链的当前结构参数，以更新第一氨基酸链的当前结构参数。
Need to check novelty before this filing date? Find Prior Art