📄 中文摘要
蛋白质二级结构预测对于理解蛋白质功能和推动药物发现至关重要。然而，复杂的序列-结构关系给准确建模带来了显著挑战。为了解决这些问题，提出了一种多目标遗传编程框架MOGP-MMF，将蛋白质二级结构预测重新构建为一个聚焦于特征选择和融合的自动优化任务。MOGP-MMF引入了一种多视角多层次表示策略，整合了进化视角、语义视角和新引入的结构视角，以捕捉全面的蛋白质折叠逻辑。通过丰富的操作符集，该框架能够演化线性和非线性融合函数，有效捕捉高维特征之间的复杂关系。
基于多目标遗传编程和多视角多层次特征的增强蛋白质二级结构预测
📄 English Summary
Multi-objective Genetic Programming with Multi-view Multi-level Feature for Enhanced Protein Secondary Structure Prediction
Predicting protein secondary structure is crucial for understanding protein function and advancing drug discovery. The intricate sequence-structure relationship presents significant challenges for accurate modeling. To tackle these issues, a multi-objective genetic programming framework, MOGP-MMF, is proposed, reformulating protein secondary structure prediction as an automated optimization task centered on feature selection and fusion. MOGP-MMF introduces a multi-view multi-level representation strategy that integrates evolutionary, semantic, and newly introduced structural views to capture the comprehensive logic of protein folding. Utilizing an enriched operator set, the framework evolves both linear and nonlinear fusion functions, effectively capturing the complex relationships among high-dimensional features.