Additional file 1 of QSAR-derived affinity fingerprints (part 1): fingerprint construction and modeling performance for similarity searching, bioactivity classification and scaffold hopping

Additional file 1. QAFFP construction. Statistics (ligand count and quality criteria $$q^{2}$$ q 2 and $$R_{0}^{'2}$$ R 0 ′ 2 ) of all 1360 trained models. Each model is given by the target ChEMBL ID and by an activity type (Ki, Kd, IC50 or EC50). 440 models used to construct the QAFFP fingerprint highlighted in green. Ligand data of 440 models (given as ligand ChEMBL ID, SMILES, activity type and activity value).

补充材料1. QAFFP构建流程。本部分涵盖全部1360个训练模型的统计信息，具体包括配体数量以及质量准则$q^2$与$R'^2_0$。每个模型均通过对应的靶标ChEMBL ID以及活性类型（抑制常数Ki、解离常数Kd、半最大抑制浓度IC50或半最大效应浓度EC50）进行标识。其中，用于构建本次重点展示的QAFFP指纹的440个模型已以绿色高亮标出。此外还提供了这440个模型的配体数据，具体包含配体ChEMBL ID、简化分子线性输入规范（SMILES）、活性类型以及活性数值。