Training Nmers randomly subsampled to reduce expression bias

Mutated Nmers from the balanced training set described in the manuscript titled "A machine learning model for ranking candidate HLA class I neoantigens based on known neoepitopes from multiple human tumor types". These Nmers have been randomly selected to reduce the expression bias in the data amongst the screened nmers本数据集取自题为《基于多个人类肿瘤类型已知新表位的HLA I类（HLA class I）新抗原候选物排序机器学习模型》的研究论文中描述的平衡训练集，包含其中的突变型N聚体（Mutated Nmers）。此类N聚体经随机选取，以降低筛选所得N聚体相关数据中的表达偏倚。