Diverse Docking Benchmark Set

[drewnutt]

Are you able to provide the full set of data used for the docking benchmark?

From my understanding of the paper, this would consist of:

• a list of the 245 unique ligands and which of the 30 target proteins they are docked to
  • Are these the ligands present in the PDB structures enumerated in stats_data/pdbs.txt?
• the helper ligands from Chembl for each of the 245 ligands (SMILES strings would be great)
• denote which helper ligands are utilized when focusing on one of the two criteria: 1) highest affinity binders that do not share a chemical scaffold 2) ligands that share the largest chemical substructure with the query ligand

[jpaggi]

Please see the files "stats_data/pdbs_for_benchmark.csv", "stats_data/helper_best_affinity_diverse.csv", and "stats_data/helper_best_mcss.csv" and the descriptions I added to the readme under the header "Benchmarking data"

The first file lists the ligands used in our benchmarks and the structure to which they are docked. The 248 ligands shown in figure 4 of our paper are those for which both the "any_near_native" column and the "mcss<0.5" column are "True".

The two remaining files provide the chembl ligands used under each selection criteria (smiles strings included).