Sample Docking Report: EGFR and Erlotinib (1M17)

Result at a glance

1.94 ÅRMSD of the top pose to the crystal pose

−6.9Best Vina score (kcal/mol)

2 of 10Poses within 2 Å of the crystal

KeptHinge hydrogen bond to Met769

The top scoring pose lands 1.94 Å from the crystal pose, inside the usual 2 Å limit for a successful redocking, and pose 2 is closer still at 1.67 Å. In other words, ChemServer's default settings put erlotinib back where the crystal structure has it.

Docked and crystal pose in 3D

Green sticks: docked pose 1. Grey sticks: erlotinib as found in the crystal. The hinge residue Met769 is labelled. Drag to rotate, scroll or pinch to zoom.

The target

The kinase domain of EGFR (epidermal growth factor receptor) bound to erlotinib, an approved drug for non small cell lung cancer. Structure 1M17 was solved at 2.6 Å by Stamos, Sliwkowski and Eigenbrot (J Biol Chem, 2002). Erlotinib (ligand code AQ4) sits in the ATP pocket and is held by a hydrogen bond from the hinge residue Met769. It is one of the classic test cases for checking a docking setup.

What was run

Structure	PDB 1M17, fetched from RCSB
Receptor	Water and other hetero atoms removed, then AutoDockTools 1.5.7 (polar hydrogens, Gasteiger charges)
Ligand	Erlotinib cut out of 1M17 as a PDB file (29 heavy atoms) and uploaded like any user file. Open Babel rebuilt it in 3D with hydrogens and Gasteiger charges, so the docking did not start from the crystal shape.
Search box	Reference ligand mode: centred on the crystal ligand at (21.7, 0.3, 52.1), 34 × 30 × 30 Å
Docking	AutoDock Vina 1.2.5, exhaustiveness 32, 10 poses, energy range 4 kcal/mol
Settings	ChemServer's defaults. Nothing was tuned for this target.

All poses

Pose	Vina score (kcal/mol)	RMSD to crystal (Å)
1	−6.9	1.94
2	−6.9	1.67
3	−6.7	2.39
4	−6.5	8.11
5	−6.5	4.33
6	−6.4	7.12
7	−6.4	8.33
8	−6.4	3.11
9	−6.4	7.69
10	−6.3	2.82

RMSD is measured over the 29 heavy atoms of erlotinib, in place without superposing the two poses, and counts symmetric atoms correctly. Below 2 Å is the usual sign of a correct pose.

Same input on every engine

The same structure and ligand file, run through each docking engine ChemServer offers. Vina-GPU searches at most 30 Å per side, so for the two Vina-GPU runs ChemServer trimmed the box around erlotinib from 34 × 30 × 30 to 30 × 30 × 30 Å. Scores from different engines are not on the same scale, so compare the RMSD columns. Run times are for the whole job, preparation included.

Engine	Top pose score (kcal/mol)	Top pose RMSD (Å)	Closest pose RMSD (Å)	Run time
AutoDock Vina 1.2.5 (CPU)	−6.9	1.94	1.67 (pose 2)	65 s (compute server CPU)
Vina-GPU 2.1 (GPU)	−6.7	1.71	1.56 (pose 2)	9 s (compute server GPU)
GNINA 1.3 (GPU, CNN scoring)	−7.2	1.48	1.20 (pose 2)	76 s (compute server GPU)
Vina-GPU 2.1 + GNINA CNN (GPU)	−6.5	1.78	1.44 (pose 5)	32 s (compute server GPU)

When no compute server is connected, ChemServer runs the job on the web server itself, more slowly. The same AutoDock Vina job run that way gave the same answer, a top pose at −6.9 kcal/mol and 1.92 Å RMSD, in about 15 minutes.

Interactions: crystal versus docked

Both the crystal pose and the docked pose keep the key hydrogen bond from the backbone NH of Met769 in the hinge to the quinazoline nitrogen of erlotinib. Of the 6 interactions found, 3 appear in both poses. Interactions were detected with ProLIF after adding hydrogens, the same way the 2D interaction map in the app works.

Residue (chain A)	Interaction	Crystal	Docked
Val702	Hydrophobic contact	No	Yes
Lys721	Hydrophobic contact	Yes	Yes
Met742	Hydrophobic contact	No	Yes
Leu764	Hydrophobic contact	Yes	Yes
Met769	Hydrogen bond (ligand accepts)	Yes	Yes
Asp831	Hydrophobic contact	Yes	No

What you get in the download

The results ZIP holds exactly what the app gives you after a run (13 files):

docking_results.xlsx with the best score per compound, every pose and the full summary
docking_summary.csv with the same summary as plain text
METHODS_AND_CITATIONS.txt with the methods text and references shown below
config.txt with the Vina box and settings, to repeat the run on your own computer
Receptor and ligand files as prepared (PDB and PDBQT) and the docked poses (PDBQT)
Logs of every step

Methods text written for this run

Every run comes with this file, ready to adapt for a methods section.

Methods used for this run
==========================
Docking engine: AutoDock Vina
Structure preparation: AutoDockTools 1.5.7 (MGLTools) prepare_receptor4.py for the receptor (polar hydrogens, Gasteiger charges); ligands built in 3D and charged (Gasteiger) with Open Babel.
Software versions: Open Babel Open Babel 3.1.1 -- Mar 31 2024 -- 06:39:03; AutoDock Vina AutoDock Vina v1.2.5; Meeko 0.8.0; RDKit 2026.09.1; MGLTools 1.5.7 (native)

Please cite:
1. Eberhardt J, Santos-Martins D, Tillack AF, Forli S. AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings. J Chem Inf Model. 2021;61(8):3891-3898. doi:10.1021/acs.jcim.1c00203
2. Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem. 2010;31(2):455-461. doi:10.1002/jcc.21334
3. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, Olson AJ. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J Comput Chem. 2009;30(16):2785-2791. doi:10.1002/jcc.21256
4. O'Boyle NM, Banck M, James CA, Morley C, Vandermeersch T, Hutchison GR. Open Babel: An open chemical toolbox. J Cheminform. 2011;3:33. doi:10.1186/1758-2946-3-33

Repeat this run yourself

Enter the engine and open Docking.
Fetch 1M17 by its PDB ID.
Upload erlotinib_from_1M17.pdb as the ligand.
Keep Reference Ligand (Auto) and choose AQ4 as the reference ligand.
Press start. The pose viewer shows the same RMSD check under Protocol Validation.

Sample docking report: EGFR with erlotinib