shrishdwivedi/vmap-dataset

ViralMap数据集是一个用于病毒蛋白质特征预测的数据集，基于主要序列数据。它包含一个Pandas DataFrame文件（vmap_data.pkl），其中存储了序列、标签、簇ID、折叠分配、分类学信息以及其他UniProt相关列；此外，还包括测试折叠结果文件夹（test_fold{1-5}/），这些文件夹中包含来自交叉验证的测试折叠结果，每个测试折叠蛋白质对应一个CSV文件。该数据集旨在支持从初级序列预测病毒蛋白质特征的研究，如通过机器学习或生物信息学方法进行分析。

The ViralMap dataset is designed for predicting features in viral proteins from primary sequence data. It includes a Pandas DataFrame file (vmap_data.pkl) containing sequences, labels, cluster IDs, fold assignments, taxonomy, and other UniProt columns; additionally, it provides test fold results from cross-validation in folders (test_fold{1-5}/), with one CSV file per test fold protein. This dataset supports research on feature prediction in viral proteins using methods such as machine learning or bioinformatics.