Re: [AMBER] How to make parameter files of GTP from PDB file.

From: Bill Ross <ross.cgl.ucsf.EDU>
Date: Thu, 5 Aug 2010 10:28:45 -0700

> Step-1: preparing GTP-solution
>
> >source leaprc.ff99SB
> >loadamberprep GTP.prep
> >loadamberparams frcmod.phos
>
> >x = gtp

Normally one would do 'x = copy gtp' to keep the gtp template itself
clear of water, but if you don't use gtp again in the session, it
should not matter.

> >solvatebox x TIP3PBOX 10
> >savepdb x solute.pdb

> Step-2: preparing parameter files with specific water model

> >source leaprc.ff99SB
> >WAT = SPC
> >loadAmberParams frcmod.spce
> >loadamberprep GTP.prep
> >loadamberparams frcmod.phos
> >x = loadpdb solute.pdb
> >setBox x centers
> >saveamberparm x gtp_10A_SPC.prmtop gtp_10A_SPC.prmcrd
> >quit

> Here, I used prep file and frcmod file downloaded from Amber Home Page.


> However, I encountered a problem in step-2 when loading the PDB file, "solute.pdb".

> The error message,

> Created a new atom named: O2* within residue: .R<gtp 1>
> Created a new atom named: HO*2 within residue: .R<gtp 1>
> .
> .
> .
> Added missing heavy atom: .R<gtp 1>.A<O2' 43>
> .
> .
> FATAL: Atom .R<gtp 1>.A<O2* 45> does not have a type.
> FATAL: Atom .R<gtp 1>.A<HO*2 46> does not have a type.
> .
> .

It appears that when you did the savepdb the O2' and HO'2 atoms
were mapped to O2* and HO*2, and these weren't recognized on
loadpdb. There is an 'atom name map' in the typical leaprc that
governs the conversion of atom names and should be fixable.

You can postpone this problem by not doing a savepdb followed
by a loadpdb, just set up in one leap session. However you may
face this problem later, so it does not hurt to solve it now.

Bill


> was shown and paramer files cannot be saved.

> I would like to know how to circumvent this error.


> You would think that I should do in 1 step.
> Of coures, I tried it.
> I found that I cannot
> If I use waterbox (e.g. )

> It is important for us to use waterbox to keep whole watar molecule number
> among different water model solvent.

> Thank you for kind support,

> Ikuo Kurisaki



_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Aug 05 2010 - 10:30:07 PDT
Custom Search