ADL: Ligand docking to MD ensemble

de Waal, Parker Parker.DeWaal at vai.org
Wed Apr 9 08:24:20 PDT 2014


Hi James,

I've never used gold however you could use a bash script like this to automate the entire process:

touch complete_log.dat
for r in receptor/*.pdbqt; do 
	echo "Docking $r ..."
	vina --conf config.txt --receptor $r --out $r.all --log $r.dat
	cat $r.dat >> complete_log.dat
done

Additionally here is a sample config file:
ligand = JZN.pdbqt

center_x =  59.922
center_y =  4.948
center_z =  0

size_x = 30
size_y = 22
size_z = 36

Regarding your issue with atom types: What are you using to convert your trajectory -> pdb -> pdbqt? I don't have much experience with CHARMM as I use AMBER, however I have never experienced this problem. Are you using AutoDock Tools to make the pdbqt file?

Best,
Parker
-----Original Message-----
From: autodock-bounces at scripps.edu [mailto:autodock-bounces at scripps.edu] On Behalf Of James Starlight
Sent: Wednesday, April 09, 2014 10:33 AM
To: autodock at scripps.edu
Subject: Re: ADL: Ligand docking to MD ensemble

Parker,

thanks for suggestions! But are there any possibilities to define ensemble instead of single receptor's conformation as dcd or in nrm-like format in Vina's conf file (I've seen such in GOLD) ?

By the way also I've forced with the problem during processing of my receptor.pdbqt. I've made pdbqt from the charmm-format pdb using ADT but after processing conf file in Vina I've received error something like "Atomtype CE not found" How I can convert all atom names to autodock-like format ?

James


2014-04-09 16:40 GMT+04:00 de Waal, Parker <Parker.DeWaal at vai.org>:

> Ahh,
>
> I misunderstood the question. I suppose to identify high affinity 
> conformations you could start by taking evenly spaced snapshots of 
> your simulation every 100 ps or so. Then You could dock your ligand 
> into each conformation and determine which conformations have a 
> relative binding affinity. Using this you could then try to determine 
> the characteristics of those snapshots with higher binding affinities 
> and how they differ from snapshots with lower binding affinities and perform some PCA to quantify.
> After you could search through your trajectory for frames with similar 
> characteristics to those determined by your PCA and you should be able 
> to narrow down your search.
>
> Alternatively... starting from your simulated apo protein I would 
> simply dock your ligand into the binding site and run a ligand bound simulation.
> That would probably provide you with a more accurate answer than a 
> static binding.
>
> Best,
> Parker
>
> -----Original Message-----
> From: autodock-bounces at scripps.edu 
> [mailto:autodock-bounces at scripps.edu]
> On Behalf Of James Starlight
> Sent: Wednesday, April 09, 2014 8:30 AM
> To: autodock at scripps.edu
> Subject: Re: ADL: Ligand docking to MD ensemble
>
> Hi Parker,
>
> I think I'm not need in MMPBSA because the only I need to dock my 
> ligand to one of the conformation obtained as the result of the apo 
> protein simulation. E.g random selected conformation from apo 
> trajectory can be with the locked active site if the docking will be 
> accompanied with the sterical clashes :) So i need simple algorithm of 
> the selection the most
> (sterically) appropriated (defining active site) conformation from the 
> ensemble.
>
>
> James
>
>
> 2014-04-09 16:22 GMT+04:00 de Waal, Parker <Parker.DeWaal at vai.org>:
>
> > Hi James,
> >
> > You might want to look into MMPBSA or MMGBSA calculations. Using 
> > these methods you can compute relative binding energies for various 
> > conformations throughout your simulation which will probably be more 
> > accurate than the internal auto dock calculations if configured
> correctly.
> >
> > Best,
> > Parker
> >
> > -----Original Message-----
> > From: autodock-bounces at scripps.edu
> > [mailto:autodock-bounces at scripps.edu]
> > On Behalf Of James Starlight
> > Sent: Wednesday, April 09, 2014 8:16 AM
> > To: autodock at scripps.edu
> > Subject: ADL: Ligand docking to MD ensemble
> >
> > Dear AutoDock users!
> >
> >
> > I'm looking for the Autodock tutorial where I could  find brief 
> > description of how protein flexibility from the MD simulation could 
> > be take into account during the docking run. For instance I have 
> > ligand and ensemble of conformations  in dcd or pdb format 
> > corresponded to the fragment of my MD trajectory. Now I'd like to 
> > dock my ligand to each of that conformation in its pre-defined 
> > active site and find to what conformation from ensemble my ligand 
> > has higher affinity. I'll be thankful for any suggestions.
> >
> > Thanks for suggestions,
> >
> > James
> > ________________________________________________
> > --- ADL: AutoDock List  ---
> > http://scanmail.trustwave.com/?c=129&d=jb3F08Zq7bBnucop847eM5SvGqYRi
> > ON
> > 1hVYI4wLGVQ&u=http%3a%2f%2fautodock%2escripps%2eedu%2fmailing%5flist
> > --
> > -
> >
> > ________________________________________________
> > --- ADL: AutoDock List  ---
> > http://scanmail.trustwave.com/?c=129&d=jb3F08Zq7bBnucop847eM5SvGqYRi
> > ON 
> > 1hQFc5FnBDQ&u=http%3a%2f%2fautodock%2escripps%2eedu%2fmailing%5flist
> > ---
> >
> ________________________________________________
> --- ADL: AutoDock List  ---
> http://scanmail.trustwave.com/?c=129&d=m9rF0ye3Fjl07MG9eKJv03fPJ6MFJ_M
> XdHgYuG82yA&u=http%3a%2f%2fautodock%2escripps%2eedu%2fmailing%5flist--
> -
>
> ________________________________________________
> --- ADL: AutoDock List  --- 
> http://scanmail.trustwave.com/?c=129&d=m9rF0ye3Fjl07MG9eKJv03fPJ6MFJ_M
> XdC9MvzQxkA&u=http%3a%2f%2fautodock%2escripps%2eedu%2fmailing%5flist 
> ---
>
________________________________________________
--- ADL: AutoDock List  --- http://scanmail.trustwave.com/?c=129&d=m9rF0ye3Fjl07MG9eKJv03fPJ6MFJ_MXdC9MvzQxkA&u=http%3a%2f%2fautodock%2escripps%2eedu%2fmailing%5flist ---



More information about the autodock mailing list