MDDTA: A drug target binding affinity prediction method based on molecular dynamics simulation data enhancement.

MD-PDBbind.zip:Dynamic database for training, containing the natural conformation of each complex and the corresponding 10-frame dynamic conformation. PDBbind_2020_md.csv: Enhanced MD-PDBbind sample set used for training. CASF2016.csv: CASF-2016 test set. protein_emb_data.h5: ESM2 pre-training characterization file containing proteins from both the training and test sets. protein_coords_data.h5: File containing protein pocket coordinates for both the training and test sets. ligand_data.h5: File containing the coordinates of ligand small molecules and GraphMVP pre-training characterization for both the training and test sets. drugbank_compound.zip: Compound PDBQT structure files required for Qvina-W docking in drugbank drug screening. Drugbank_dataset.zip: The dataset required for MDDTA prediction in drugbank drug screening and the h5 file with embedded characterisation and coordinates.MD-PDBbind.zip：用于训练的动态数据库，包含各复合物的天然构象及其对应的10帧动态构象。 PDBbind_2020_md.csv：用于训练的增强版MD-PDBbind样本集。 CASF2016.csv：CASF-2016测试集。 protein_emb_data.h5：包含训练集与测试集蛋白质的ESM2预训练表征文件。 protein_coords_data.h5：包含训练集与测试集蛋白质口袋坐标的文件。 ligand_data.h5：包含训练集与测试集配体小分子坐标以及GraphMVP预训练表征的文件。 drugbank_compound.zip：用于DrugBank药物筛选中Qvina-W分子对接所需的化合物PDBQT结构文件。 Drugbank_dataset.zip：用于DrugBank药物筛选中MDDTA预测所需的数据集，以及包含表征与坐标的h5文件。