Re: Is it possible to rename atoms with psfgen?

From: Brian Radak (brian.radak.accts_at_gmail.com)
Date: Fri Jan 30 2015 - 10:35:43 CST

Thanks Jim!

After a little goofing around, that works. However, I'll note the
following. I'm not sure if this is expected or avoidable behavior:

The original topology has two side chains that are exact one-to-one copies.
I arbitrarily distinguished these by adding an extra "1" to the name of the
copies. the following selections give apparently harmless errors:

# This is meant to only return the copied atoms, i.e. the ones with
non-standard names that VMD apparently
# recognizes should not be there.
set l1atoms [atomselect top "segid PROT and resid 2 and not sidechain and
not backbone"]

ERROR) atomselection: sidechain: protein residue index 1, C-alpha index 3
has more than two non-backbone bonds; ignoring the others

# This is meant to select everything that is not the original sidechain,
i.e. replace that side chain with the
# copied atoms. Note that those atoms were already selected and renamed at
this point in the script. It
# appears that VMD searches the sidechains in order based on connectivity
and recognizes the first one only,
# even though the second one has the standard names at this point.
set newsystem [atomselect top "not (segid $SEGID and resid $RESID and
sidechain)"]

ERROR) atomselection: sidechain: protein residue index 1, C-alpha index 3
has more than two non-backbone bonds; ignoring the others

Brian

On Thu, Jan 29, 2015 at 1:11 PM, Jim Phillips <jim_at_ks.uiuc.edu> wrote:

> Hi,
>
> Atom renaming isn't supported in psfgen, but I think you can do your
> entire "unpatching" process more cleanly in the latest VMD, which can now
> write psf files containing full bond/angle/dihedral/improper/... data
> from an atom selection. In VMD you can rename atoms something something
> like:
>
> [atomselect top "resname FOO and name BARX BARY"] set name BAR
>
> Then to write out:
>
> set a [atomselect top "...lambda 0 atoms..."]
> $a writepsf a_lam0.psf
> $a writepdb a_lam0.pdb
>
> If you want to do this to the output trajectory then use catdcd.
>
> Jim
>
>
>
> On Wed, 28 Jan 2015, Brian Radak wrote:
>
> I'm working on a somewhat complicated workflow utilizing psfgen and
>> combinations of single and dual topologies.
>>
>> I can successfully apply a patch with psfgen to convert a single topology
>> psf into a dual topology psf. I can also "unpatch" the dual topology psf
>> and keep either set of the copied atoms (what you might call the "lambda
>> 0"
>> and "lambda 1" sets). Obviously the lambda 0 and lambda 1 atoms have
>> different names. However, I'd like the resulting "unpatched" single
>> topology psf to always have the same standard pdb names (regardless of
>> which atoms I kept).
>>
>> In CHARMM I would have supplemented the patch with the "RENAME ATOM"
>> command, but I can't readily find an analogous command in psfgen, although
>> pdbalias sounds very close.
>>
>> Can anyone make a suggestion? I suppose, although this would be very
>> kludgy, that I could just use a system call to sed and change the names
>> that way.
>>
>> Thanks,
>> Brian
>>
>> --
>> Brian Radak
>> Postdoctoral Scholar
>> University of Chicago
>> Department of Biochemistry & Molecular Biology
>> Gordon Center for Integrative Science, W323A
>> 929 E. 57th St.
>> Chicago, IL 60637-1454
>> Tel: 773/834-2812
>> e-mail: radak_at_uchicago.edu
>>
>>

-- 
Brian Radak
Postdoctoral Scholar
University of Chicago
Department of Biochemistry & Molecular Biology
Gordon Center for Integrative Science, W323A
929 E. 57th St.
Chicago, IL 60637-1454
Tel: 773/834-2812
e-mail: radak_at_uchicago.edu

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:21:37 CST