Test set of 140 complexes for AutoDock-GPU

Set of 140 protein-ligand complexes<br> ===================================<br> <br> # Overview<br> <br> The ligands herein vary in the number of atoms and number of rotatable bonds.<br> This is the full data set used in the following study:<br> Accelerating AutoDock4 with GPUs and Gradient-Based Local Search<br> https://dx.doi.org/10.26434/chemrxiv.9702389.v1<br> <br> # Warning<br> <br> These structures, both proteins and ligands, were prepared in an automated way<br> without manual inspection. The following is a non-comprehensive list of<br> issues that may exist:<br> <br> * missing water molecules that bridge ligand-receptor interactions,<br> * missing atoms in the proteins,<br> * non-integer sum of partial charges,<br> * incorrect protonation state,<br> * incorrect protein conformations.<br> <br> This intended use of this data is to evaluate the performance of docking with<br> regard to computational and algorithmic efficiency, but not the accuracy of<br> the scoring function.<br> <br> # Details<br> <br> In ligand\_properties.csv<br> * `pdb` Protein Data Bank accession code<br> * `n_atom` number of atoms in the ligand<br> * `n_tors` number of bonds in the ligand that can rotate during docking.<br> * `score_of_probable_global_minimum` lowest (best) score (score is the sum of intra- and inter-mo<br> leculer energy components). The best score did not improve with an increase in the search effort<br> * `RMSD_of_probable_global_minimum` RMSD from X-ray pose for the solution with the lowst score <br> * `best_score_so_far` lowest score ever found but even lower scores are likely to exist<br> <br>