Thanks for your answers.
There was indeed a space between the last ambmask and minimum value in my
input file, it apparently got lost when typing the first email. If I put
double quotations around each ambmask the restraint is indeed still not
recognized.
Changing the atomnames to a specific unique name without the apostrophe
indeed does the trick. On Friday, I quickly wrote the dinucleotide in my
prep-file and it's working now. This morning, I changed the atomnames in the
topology file (of the original dinucleotide made from 2 residues) and this
indeed also does the trick.
Thanks for your help
Charles-Alexandre Mattelaer
-----Original Message-----
From: David Cerutti [mailto:dscerutti.gmail.com]
Sent: vendredi 19 mai 2017 19:23
To: AMBER Mailing List <amber.ambermd.org>
Subject: Re: [AMBER] mdgx configuration sampling
It looks like mdgx is going to recognize either single or double quotes as
'quotation' and so I'm not sure what to do about making amber masks with
single quote / prime symbols in the names. You can go into the topology,
replace the prime symbols with 'p' or some other innocuous letter, and then
proceed. Meanwhile, I'll ask around to see how Fortran namelists parse
single and double quotes and get prime symbols into words--I think I can
guess how it's supposed to be done, and the ParseWords function in mdgx
Parse.c shouldn't be hard to fix.
Dave
On Fri, May 19, 2017 at 11:33 AM, David Cerutti <dscerutti.gmail.com> wrote:
> The "Configs >> Processed 0 restraint operations" output means that
> it didn't recognize your restraint command. Very likely the
> apostrophes in the atom names are the problem--it is thinking that
> there is a quotation in there perhaps. First, make sure that there is a
space between ":2.C5'
> <space> min" and let me have a look at the code for processing
> keywords with apostrophes. I know I've programmed it to be able to
> detect them in some fashion but I also need to reserve double and
> single quotations to demarcate phrases (like ambmasks) that should all be
taken as one input.
> Try putting additional, double quotations around each of the ambmasks
> like
> this:
>
> GridSample ":2.P" ":2.O6'" ":2.C6'" ":2.C5'" min -180 max 180 Krst 64
>
> I'll have a look in the code to see if there's some other convention
> you can use.
>
> Dave
>
>
> On Fri, May 19, 2017 at 7:55 AM, CA gmail <camattelaer01.gmail.com> wrote:
>
>> Dear Amber users
>>
>>
>>
>>
>>
>> I'm using mdgx from AmberTools 17 for configurational sampling of a
>> backbone in a (modified) dinucleotide molecule. However, apparently
>> mdgx doesn't recognize atoms, involved in torsions/angles etc., in
>> another residue than the first. For example, this input file:
>>
>>
>>
>> &files
>>
>> -p ../AC.prmtop
>>
>> -c ../AC.inpcrd
>>
>> -o GenConformers.out
>>
>> &end
>>
>>
>>
>> &configs
>>
>> GridSample :2.P :2.O6' :2.C6' :2.C5'min -180 max 180 Krst 64
>>
>> maxcyc 10
>>
>> count 10
>>
>> verbose 1
>>
>> outbase 'Conf'
>>
>> write 'pdb'
>>
>> outsuff 'pdb'
>>
>> &end
>>
>>
>>
>> Gives following statement as output:
>>
>>
>>
>> Configs >> created 10 initial configurations
>>
>> Configs >> Processed 0 restraint operations
>>
>> . (it does complete normally)
>>
>>
>>
>> And outputs 10 identical configurations. I can additionally add that
>> removing the ':2' (there is only one possible combination of those 4
>> atomnames) doesn't help Can this possibly be a bug or is mdgx not
>> meant for molecules consisting of multiple 'building blocks'? Should
>> I write a new entry in my prep-file for the dinucleotide or could I
>> alter the .prmtop-file in a way mdgx would be able to identify the
>> necessary atoms?
>>
>>
>>
>> Kind regards and thanks in advance
>>
>>
>>
>> Charles-Alexandre Mattelaer
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon May 22 2017 - 01:00:02 PDT
