通过数据驱动方法预测蛋白质聚集倾向
Prediction of Protein Aggregation Propensity via Data-driven Approaches
AI总结：
本研究利用两类数据库构建数据驱动模型预测蛋白质聚集倾向，开发了基于特征和图的模型，发现基于图的回归模型表现更优（R2为0.95），并指出有效方法取决于描述符和数据库类型。
摘要：
当错误折叠或未折叠的蛋白质发生物理结合时，会出现蛋白质聚集，并可能促进各种淀粉样疾病的发展。本研究旨在利用两类数据库，通过数据驱动方法构建用于预测蛋白质聚集的代理模型。首先，使用Aggrescan3D 2.0计算Protein Data Bank中蛋白质结构的分数，构建了聚集倾向分数数据库。此外，利用该数据库开发了基于特征和基于图的蛋白质聚集预测模型。基于图的回归模型优于基于特征的模型，得出0.95的R2，尽管其本质上需要蛋白质结构。其次，对于实验数据，使用Curated Protein Aggregation Database 2.0构建了基于特征的模型，以预测聚集强度曲线。总之，本研究提出了更有效预测蛋白质聚集的方法，具体取决于描述符和数据库的类型。
英文摘要：
Protein aggregation occurs when misfolded or unfolded proteins physically bind together, and can promote the development of various amyloid diseases. This study aimed to construct surrogate models for predicting protein aggregation via data-driven methods using two types of databases. First, an aggregation propensity score database was constructed by calculating the scores for protein structures in Protein Data Bank using Aggrescan3D 2.0. Moreover, feature- and graph-based models for predicting protein aggregation have been developed using this database. The graph-based regression model outperformed the feature-based model, resulting in R2 of 0.95, although it intrinsically required protein structures. Second, for the experimental data, a feature-based model was built using Curated Protein Aggregation Database 2.0, to predict the aggregated intensity curves. In summary, this study suggests the approaches that are more effective in predicting protein aggregation, depending on the type of descriptor and the database.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
