proteinglm/temperature_stability

Temperature Stability Dataset是一个用于预测蛋白质在65摄氏度条件下结构稳定性的数据集。每个数据实例包含一个蛋白质序列字符串和一个整数标签，标签表示该序列在65摄氏度下是否能够保持结构稳定性。数据集分为训练集、验证集和测试集，分别包含283,057、62,973和73,205个实例。数据集的来源是TemStaPro，并遵循Apache-2.0许可。

The Temperature Stability Dataset is designed for predicting the structural stability of proteins at a temperature of 65 degrees Celsius. Each instance in the dataset includes a protein sequence string and an integer label indicating whether the sequence can maintain its structural stability at 65 degrees Celsius. The dataset is divided into training, validation, and test sets, containing 283,057, 62,973, and 73,205 instances respectively. The dataset is adapted from TemStaPro and is released under the Apache-2.0 License.