. Author manuscript; available in PMC: 2020 Jan 1.

Published in final edited form as: J Comput Aided Mol Des. 2018 Jul 10;33(1):19–34. doi: 10.1007/s10822-018-0133-y

Table 1:

Targets that appeared in at least one of the subchallenges of D3R Grand Challenge 3, with the PDB IDs used as references for docking; the most similar target in the PDBbind 2016 refined set, used to train the CNN, with its similarity to the provided FASTA sequence for the D3R target; mean and maximum Tanimoto coefficient of the crystal ligand associated with the PDBbind refined set target to the compounds in the challenge.

Target	reference PDB IDs	PDBbind/similarity	mean/max Tanimoto
CatS	2F1G, 2HXZ 2G7Y,3KWN 2HHN	2HHN/0.991	0.209/0.256
JAK2(SC2)	2W1I, 3E62 3JY9, 3UGC 4AGC, 5I4N 5UT2, 5UT5 5UT6	4JIA/0.980	0.291/0.437
VEGFR2	1VR2, 1YWN 2OH4, 2P2H 2P2I, 2GU5 3B8R, 3VNT	4ASE/0.657	0.214/0.414
p38α	1M7Q, 1OVE 1W82, 1W83 1WBS, 2GHL 2ZB1, 3ITZ 3L8S, 3NNU	1YQJ/0.978	0.278/0.488
JAK2(SC3)	2W1I, 3E62 3JY9, 3UGC 4AGC, 5I4N 5UT2, 5UT5 5UT6	4JIA/0.980	0.357/0.413
TIE2	2OO8, 2OSC 2P4I, 2WQB 3L8P, 4X3J	4V01/0.482	0.239/0.363
ABL1	1FPU, 1OPJ 2G1T,2G2F 2G2H,2G2I 2V7A	3K5V/0.979	0.302/0.332