SaProtHub/Dataset-RASH_HUMAN

该数据集包含人类RASH蛋白的单点突变及其通过深度突变扫描实验得到的相应突变效应分数。蛋白质序列以氨基酸序列格式给出。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。标签代表每个蛋白质的突变效应分数，分数范围从负无穷到正无穷，0代表野生型的分数，分数越高表示蛋白质的适应性越强。

This dataset contains single-site mutations of the human RASH protein and their corresponding mutation effect scores obtained from deep mutation scanning experiments. The protein sequences are provided in the amino acid sequence format. The dataset is split into training, validation, and test sets, containing 2479, 338, and 317 samples respectively. The label represents the mutation effect score for each protein, ranging from negative infinity to positive infinity, with 0 being the score of the wildtype, and higher scores indicating greater fitness of the protein.