hugging-science/shehata-antibody-psr

该数据集包含398个人类抗体重链可变域（VH）序列，带有PSR（多特异性试剂）测量值，根据Sakhnini等人（2025年）的方法进行了预处理。原始数据集由Shehata等人（2019年）发布，用于研究亲和力成熟与抗体特异性之间的关系。这是一个预处理版本，用作测试集，用于评估跨实验转移学习（ELISA训练模型→PSR测试数据）。数据集高度不平衡，仅有7个高PSR序列（1.8%）。关键特征包括：人类来源、抗体重链可变域、来自健康供体的B细胞、PSR流式细胞术检测、二进制分类标签（0=低PSR，1=高PSR）、ANARCI注释和IMGT编号方案。

This dataset contains 398 human antibody heavy chain variable domain (VH) sequences with PSR (Poly-Specificity Reagent) measurements, preprocessed according to the methodology described in Sakhnini et al. 2025 (Novo Nordisk & University of Cambridge). The dataset was originally published by Shehata et al. 2019 and contains human B cell-derived antibodies studying the relationship between affinity maturation and antibody specificity. This is the preprocessed version used as a test set for evaluating cross-assay transfer learning (ELISA-trained model → PSR test data). The dataset is highly imbalanced with only 7 high-PSR sequences (1.8%). Key features include: human origin, antibody heavy chain variable domain, B cells from healthy donors, PSR flow cytometry assay, binary classification labels (0=low PSR, 1=high PSR), ANARCI annotation with IMGT numbering scheme.