Refined AI-predicted Models with Ligands

This data record contains the energy minimized (relaxed) ligand-containing structures which are included in the relaxed AI-predicted models with ligands, but their refined version. Refinement includes, as described in the Methods section, the removal of low confidence regions with a pLDDT score below 70, and filtering for models with 10 or more amino-acid residues remaining. After refinement, ligands which are no longer within 5 Å of any atom in the model were removed. The included subfolders contain, as individual PDB files, 14,973 ligand-containing relaxed and refined AlphaFold 2 models, 15,070 ligand-containing relaxed and refined OpenFold models, and 12,855 ligand-containing relaxed and refined ESMFold models.

本数据集收录经能量最小化（弛豫，relaxed）的含配体（ligand）结构，此类结构来自带配体的经弛豫AI预测模型，且为该类模型的精修版本。精修流程如方法章节所述：移除pLDDT（predicted local distance difference test）评分低于70的低置信度区域，并过滤出剩余氨基酸残基数不少于10的模型。精修完成后，与模型内任意原子间距超出5埃（Å）的配体将被移除。所包含的子文件夹以独立PDB（Protein Data Bank）文件形式，收录了14973个经弛豫与精修的含配体AlphaFold 2模型、15070个经弛豫与精修的含配体OpenFold模型，以及12855个经弛豫与精修的含配体ESMFold模型。