SeprotHub/Dataset-RASH_HUMAN

该数据集包含了人类RASH蛋白质单点突变及其对应的突变效应分数，这些分数来自于深度突变扫描实验。蛋白质的格式为氨基酸序列。数据集分为训练集、验证集和测试集，分别包含2479、338和317个样本。标签表示基于深度突变扫描实验的每个蛋白质的突变效应分数，范围从负无穷大到正无穷大，其中0代表野生型的值，数值越大表示适应性越高。

This dataset contains single site mutations of human protein RASH_HUMAN and their corresponding mutation effect scores from deep mutation scanning experiments. The format of the proteins is amino acid sequences. The dataset is split into training, validation, and test sets with 2479, 338, and 317 samples respectively. The labels represent the mutation effect scores of each protein based on deep mutation scanning experiments, ranging from negative infinity to positive infinity, where 0 is the value of the wildtype, and larger values indicate higher fitness.