AMBER Archive (2008)

Subject: Re: AMBER: Combine mdcrd while stripping WAT problem

From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Thu Jun 12 2008 - 02:09:28 CDT

Hi:
That is an excellent suggestion. Actually, I had already tried that,
with both ambpdb and ambmask. Failure in both cases, however. I just
describe here what happened with the first. Maybe someone is
interested.

$AMBERHOME/exe/ambpdb -p complex_pop_box.prmtop <
combined_production.mdcrd > combined_production.pdb

New format PARM file being parsed
Version 1.000 Date 03/25/08
forrtl: severe (64): input conversion error, unit -5, file Internal
Formatted read

To rule out that it was a problem of shortage of memory, I run the
same for one of the production chunks. Exactly the same error.

Also, I tried with an initial prmtop/inpcrd combination for both my
ligand and the protein, and both ambpdb and ambmask worked correctly,
also in Amber10. I assume that mdcrd should behave as inpcrd to this
respect.

However, as the study at hand urged to be completed, yesterday evening
I went to a debian linux machine with larger RAM. I installed VMD and
could load all frames for the whole system of complex in hydrated
membrane and box, performing clustering and getting fine RMSD vs time
for both the protein and the ligand with plugins. Like I did months
ago to circumvent a similar problem of getting matching prmtop/mdcrd.

That does not mean that I am not interested in getting a workable
prmtop for a combined mdcrd for my systems. Either I was not correctly
doing (which is the most likely possibility), or a program to get
matching prmtop/mdcrd would be most useful. I understand the problem
of convention, however. In fact, for the initial preparation of my
systems, the part taken by Chimera was most troublesome. I had to
rearrange atom names with four characters - and much more, with great
help by Eric and Elaine - in order that leap could understand them.

Thanks

francesco

On Thu, Jun 12, 2008 at 7:59 AM, Alexander Metz
<alexander_metz2000_at_yahoo.de> wrote:
> Maybe you could create a working .prmtop file by:
>
> (1) creating a .pdb file with trajout and then
> (2) creating a new .prmtop file from this .pdb file using LEaP
>
> Good luck,
>
> Alexander
>
> --
> ++++++++++++++++++++++++++++++++++++++++++++++++++
>
>
> --- Francesco Pietra <chiendarret_at_gmail.com> schrieb am Mi, 11.6.2008:
>
>> Von: Francesco Pietra <chiendarret_at_gmail.com>
>> Betreff: Re: AMBER: Combine mdcrd while stripping WAT problem
>> An: amber_at_scripps.edu
>> Datum: Mittwoch, 11. Juni 2008, 16:57
>> Hi:
>> Still unsuccessful in loading cleanly the combined
>> (stripped) mdcrd
>> files to VMD.
>> Described here is what I did, should someone be so patient
>> to check
>> for mistakes.
>>
>> I have a number of prmtop files, corresponding to the
>> various steps. I
>> have now tried the two most naked ones, from docking the
>> ligand onto
>> the protein with DOCK6.2. Here, a prmtop was for flex
>> scoring, the
>> other one from amber score in implicit medium, that is, WAT
>> only
>> appears in the prmtop as the last residue after all protein
>> residues.
>> I deleted WAT from either one of these two prmtop files and
>> tried with
>> VMD with the combined mdcrd, stripped of :WAT :POP, and
>> nobox. Result:
>> highly distorted protein and ligand at each snapshot.
>>
>> Surely there are other combinations of mdcrd/prmtop. Those
>> I tried
>> (all other prmtop files - from embedding into the lipidic
>> membrane -
>> contain WAT BOX and result from 'solvate box model#
>> TIP3Box 12.0') led
>> to similarly distorted snapshots. I assume that I missed
>> the right
>> choice of combinations. Before trying other ones, it would
>> help to
>> appreciate where the above procedure is in error.
>>
>> Thanks a lot
>> francesco pietra
>>
>> On Tue, Jun 10, 2008 at 10:03 AM, Carlos Simmerling
>> <carlos.simmerling_at_gmail.com> wrote:
>> > whether you say nobox depends on whether your new
>> > stripped prmtop (and you must make one) has a box or
>> > not. just be consistent and it should be ok.
>> > distortions come from the mdcrd and prmtop not
>> > have the same amount of data per frame, either
>> > from mismatch in natom or mismatch in box presence.
>> > calros
>> >
>> > On Tue, Jun 10, 2008 at 3:37 AM, Francesco Pietra
>> <chiendarret_at_gmail.com> wrote:
>> >> Hi:
>> >> Perhaps you also refer to discussion I took part
>> to last year for the
>> >> same problem with Amber9.
>> >>
>> >> Like at that time, docking of the ligand was now
>> carried out (DOCK6.2)
>> >> with the protein embodying a single molecule of
>> water of
>> >> crystallization. That is, the prmtop corresponding
>> to protein+ligand
>> >> before embedding into the membrane contains that
>> WAT. I already tried
>> >> the WAT&POP stripped mdcrd with that prmtop.
>> Did not work, i.e., both
>> >> the protein and the ligand looked like heavily
>> distorted at each
>> >> snapshot when loading to VMD the stripped mdcrd
>> file.
>> >>
>> >> I can try again with prmtop stripped (with text
>> editor) of that WAT,
>> >> although as carried out above it did not work.
>> Before doing that, a
>> >> question: is it meaningful to command
>> 'nobox' while stripping only WAT
>> >> (and not POP too)?
>> >>
>> >> Thanks for the suggestions.
>> >> francesco pietra
>> >>
>> >> On Sun, Jun 8, 2008 at 9:33 PM, Carlos Simmerling
>> >> <carlos.simmerling_at_gmail.com> wrote:
>> >>> also, if you strip water, you might want to
>> use the "nobox"
>> >>> flag after trajout so that box coordinate are
>> not written. by default
>> >>> they are. if you use the correct prmtop
>> corresponding to the
>> >>> stripped system, try loading the trajectory in
>> VMD using the
>> >>> coordinates with box and see if it helps.
>> there is lots of
>> >>> discussion of this in the archives.
>> >>>
>> >>> On Sun, Jun 8, 2008 at 2:58 PM, Gustavo Seabra
>> <gustavo.seabra_at_gmail.com> wrote:
>> >>>> On Sun, Jun 8, 2008 at 11:40 AM, Francesco
>> Pietra <chiendarret_at_gmail.com> wrote:
>> >>>>> With Amber10, on accumulating ns of
>> trajectory, I am facing a problem
>> >>>>> of trajectory analysis unresolved
>> since Amber9 for the same protein
>> >>>>> and environment, though for a
>> different ligand.
>> >>>>>
>> >>>>> The system is a large protein,
>> carrying a large non-peptidic ligand.
>> >>>>> It is embedded in a POP, TIP3P
>> hydrated, membrane. Everything flows
>> >>>>> correctly, 39% faster with pmemd with
>> respect to sander.MPI.
>> >>>>>
>> >>>>> What I am trying to do with ptraj is
>> combining *.mdcrd while stripping WAT
>> >>>>>
>> >>>>> While a complete action would be:
>> >>>>>
>> >>>>> trajin prod1.mdcrd.gz
>> >>>>> trajin prod2.mdcrd.gz
>> >>>>> ......................
>> >>>>> trajout prod1-#_no_wat.mdcrd nobox
>> >>>>> strip :WAT
>> >>>>> strip :POP
>> >>>>>
>> >>>>> I tried simply:
>> >>>>>
>> >>>>> trajin prod1.mdcrd.gz
>> >>>>> trajin prod2.mdcrd.gz
>> >>>>> ......................
>> >>>>> trajout prod1-#_no_wat.mdcrd
>> >>>>> strip :WAT
>> >>>>>
>> >>>>> That in view of using the *.prmtop for
>> MD and in order not to change
>> >>>>> the residue numbering.
>> >>>>>
>> >>>>> With the same *.prmtop used for MD,
>> the combined file does not load
>> >>>>> cleanly with VMD. As expected.
>> >>>>
>> >>>> What exaclty do you mean by "not load
>> cleanly"? Do you get any error
>> >>>> messages? Anyways, I'm not sure it
>> could ever load correctly, since
>> >>>> after stripping the waters the number of
>> atoms in the prmtop file is
>> >>>> different than in the prod1-#_no_wat.mdcrd
>> file.
>> >>>>
>> >>>>> I removed all WAT from *.prmtop. The
>> same problem.
>> >>>>
>> >>>> I suppose there's more to it than just
>> removing the "WAT" residues.
>> >>>> You may want to take a look into the
>> 'rdparm' utility, described
>> >>>> together with ptraj. (Check the
>> "stripwater" and then "writeparm"
>> >>>> commands).
>> >>>>
>> >>>> Gustavo.
>> >>>>
>> -----------------------------------------------------------------------
>> >>>> The AMBER Mail Reflector
>> >>>> To post, send mail to amber_at_scripps.edu
>> >>>> To unsubscribe, send "unsubscribe
>> amber" (in the *body* of the email)
>> >>>> to majordomo_at_scripps.edu
>> >>>>
>> >>>
>> >>>
>> >>>
>> >>> --
>> >>>
>> -----------------------------------------------------------------------
>> >>> The AMBER Mail Reflector
>> >>> To post, send mail to amber_at_scripps.edu
>> >>> To unsubscribe, send "unsubscribe
>> amber" (in the *body* of the email)
>> >>> to majordomo_at_scripps.edu
>> >>>
>> >>
>> >>
>> >>
>> >> --
>> >> Dr Francesco Pietra
>> >> Professor of Chemistry
>> >> Accademia Lucchese di Scienze, Lettere e Arti,
>> founded in 1594
>> >> Palazzo Ducale
>> >> 55100 Lucca (Italy)
>> >>
>> -----------------------------------------------------------------------
>> >> The AMBER Mail Reflector
>> >> To post, send mail to amber_at_scripps.edu
>> >> To unsubscribe, send "unsubscribe amber"
>> (in the *body* of the email)
>> >> to majordomo_at_scripps.edu
>> >>
>> >
>> -----------------------------------------------------------------------
>> > The AMBER Mail Reflector
>> > To post, send mail to amber_at_scripps.edu
>> > To unsubscribe, send "unsubscribe amber" (in
>> the *body* of the email)
>> > to majordomo_at_scripps.edu
>> >
>>
>>
>>
>> --
>> Dr Francesco Pietra
>> Professor of Chemistry
>> Accademia Lucchese di Scienze, Lettere e Arti, founded in
>> 1594
>> Palazzo Ducale
>> 55100 Lucca (Italy)
>> -----------------------------------------------------------------------
>> The AMBER Mail Reflector
>> To post, send mail to amber_at_scripps.edu
>> To unsubscribe, send "unsubscribe amber" (in the
>> *body* of the email)
>> to majordomo_at_scripps.edu
>
>
> __________________________________________________________
> Gesendet von Yahoo! Mail.
> Dem pfiffigeren Posteingang.
> http://de.overview.mail.yahoo.com
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> to majordomo_at_scripps.edu
> 

-- 
Dr Francesco Pietra
Professor of Chemistry
Accademia Lucchese di Scienze, Lettere e Arti, founded in 1594
Palazzo Ducale
55100 Lucca (Italy)
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu