ADL: 回复: AutoDockFR: how to prepare ligands for covalent docking?

Zhenting Gao 183310492 at
Sat Oct 10 22:20:54 PDT 2020

Dear Sanner,

Thanks for the help. Followed your suggestion, I can generate ligand pdbqt file for covalent docking with AutoDockFR now.
To do a successful covalent docking, the matching atom defition of both receptor and ligand is very critical.
It would be nice if more details are added to the covalent docking tutorial.

Best regards

------------------ 原始邮件 ------------------
发件人:                                                                                                                        "autodock"                                                                                    <sanner at;;
发送时间:&nbsp;2020年10月11日(星期天) 凌晨1:11
收件人:&nbsp;"Gao Zhenting"<zhentgpicasa at;;"autodock at"<autodock at;;

主题:&nbsp;Re: ADL: AutoDockFR: how to prepare ligands for covalent docking?


From: Gao Zhenting <zhentgpicasa at;
Date: Saturday, October 10, 2020 at 7:02 AM
To: Michel Sanner <sanner at;, "autodock at" <autodock at;
Subject: AutoDockFR: how to prepare ligands for covalent docking?

Dear Sanner,

I am learning covalent docking workflow of AutoDockFR. And I finished the the tutorial.
The ligand is provided for the tutorial, but there seems no document about how to prepare the ligand.

I noticed that 4 atoms from sidechain of cysteine is included in the ligand pdbqt file, and they are put at the head of the pdbqt file, with atom index ranging from 1~4. If I regenerate the ligand file by, then the 4 atoms will be move at the end of the pdbqt file with index ranging from 23~26. With the regenerated ligand pdbqt file, the covalent docking pose is wrong, with a positive docking score.

Yes, this is correct. The ROOT / ENDROOT block of atoms has to contain the atoms forming the covalent bond and be the root of the torsion tree because these atoms should not move as the conformation of the ligand is modified during docking.’
If you run you need to specify which atom to use as a root for the torsion tree. If nothing is specified, will pick a central atom. The root atom can be specified using -R on the command line. I recommend running a first time on the ligand, then look at the file and find the number of the atom you want to be the root and re-run with -R Num to change the torsion tree and make sure the atoms forming the covalent bond are at the root.

With the files you sent me, running -l 3c9w_ligandWithSideChain_random_prepared_covalent_out.pdbqt -o ligandWithCorrectRoot.pdbqt -R 25

should create a ligand with the correct atoms in the root


Could you help me on how to prepare the ligand file?

Best regards

Commands of my workflow:

~/tools/MGLTools-1.5.6/bin/python /home/zhentg/tools/MGLTools-1.5.6/MGLToolsPckgs/AutoDockTools/Utilities24/ -l ../../data/3c9w_ligandWithSideChain_random.pdbqt -o 3c9w_ligandWithSideChain_random_prepared.pdbqt

~/tools/ADFRsuite-1.0/bin/adfr -l 3c9w_ligandWithSideChain_random_prepared.pdbqt -t 3c9w_cov_cmdline.trg --jobName covalent -C 1 2 3 --nbRuns 8 --maxEvals 100000 -O --seed 1

~/tools/ADFRsuite-1.0/bin/adfr -l 3c9w_ligandWithSideChain_random_prepared.pdbqt -t 3c9w_cov_cmdline.trg --jobName covalent -C 26 25 24 --nbRuns 8 --maxEvals 100000 -O --seed 1
--- ADL: AutoDock List&nbsp; --- ---

More information about the autodock mailing list