一种基于条件信息的蛋白质逆折叠方法

By employing a reinforcement learning method based on conditional information, combined with a multi-objective reward function and a classifier-free guidance strategy, this study addresses the lack of integration of key functional motifs and mate information in existing technologies. This enables high-fidelity and diverse protein reverse folding, improving the success rate and effectiveness of protein design.

CN121922190BActive Publication Date: 2026-07-17BEIHANG UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIHANG UNIV
Filing Date
2026-03-26
Publication Date
2026-07-17

Smart Images

  • Figure CN121922190B_ABST
    Figure CN121922190B_ABST
Patent Text Reader

Abstract

本发明公开了一种基于条件信息的蛋白质逆折叠方法，包括基于输入的蛋白质目标结构和条件信息，通过策略网络采样输出候选蛋白质序列以及对应的对数概率，然后通过所述对数概率确定当前候选蛋白质序列的采样概率。本申请构建了多目标奖励函数，用于对候选蛋白质序列进行评估，生成奖励值，然后根据候选蛋白质序列的对数概率和对应奖励值，对策略网络的参数进行优化。本申请在高保真恢复关键基序结构的同时，可确保全局结构的稳定性；同时在保持结构保真的基础上，能够实现对结合亲和力等功能指标的协同优化。
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Protein sequence design implementation method based on multi-objective optimization

    CN111554346A

  • Method and device for training protein prediction model based on graph neural network

    CN116935952A