Re: Getting extra atoms by Patching!

From: Kenno Vanommeslaeghe (kvanomme_at_rx.umaryland.edu)
Date: Mon May 05 2014 - 11:45:48 CDT

Hi Sadegh,

At first glance, I see nothing wrong with your patch at the *technical*
level, but it's hard to debug these things just by looking at them. It's a
bit like debugging a program; you have to carefully look at how things go
wrong, look for warning and error messages throughout the process of
generating your model, and piece back what could have happened. One
possibility is that you might have specified the residues in the wrong
order. Another possibility is that the atom *names* you used in your patch
don't correspond to the names of the same atoms in the pdb.

At the *physical* level, your atom types and charges are wrong. An amide
bond has chemically distinct properties and cannot remotely be
approximated by smashing a carboxylic acid and an amine together. Please
use standard CHARMM amide atom types and charges. Doing so might actually
fix your problem; it is possible that your patch failed because there are
no parameters for your unphysical combination of atom types.

Once you're done fixing the atom types and charges, feel free to post your
corrected patch so that I can confirm you got the atom types and charges
right. And good luck with the technical part.

        Kenno.

On 05/03/2014 02:11 AM, Michelle Kuttel wrote:
> Hi Sadegh
>
> Yes, a patch will generate atoms according to the templates in the
> topology file (in your case ASP, LYS and LINK). If the extra atoms are
> there in the original residues (ASP and LYS) and are not deleted in your
> patch, they will remain. The pdb file is used for coordinates, not as
> the rule/guide for the number of atoms.
> regards
> -----------------------------
> Michelle Kuttel
> -----------------------------
>
>
>
>
> On 03 May 2014, at 1:51 AM, Sadegh Faramarzi Ganjabad
> <safaramarziganjabad_at_mix.wvu.edu <mailto:safaramarziganjabad_at_mix.wvu.edu>>
> wrote:
>
>> Hello
>>
>> I use patch to link ASP to LYS. However, after generating psf and pdb
>> files I see 5 extra new atoms in the center of coordinate which belong
>> to ASP residue, but in fact they don't belong to it in original pdb
>> file. Has anybody faced with such a problem before?
>>
>> Thank you in advance
>> Sadegh
>>
>> Following is what I put in topology file;
>> PRES LINK 0.00 ! linkage for IMAGES or for joining segments
>> ! 1 refers to previous (N terminal) --> Asp3
>> ! 2 refers to next (C terminal) --> Lys10
>> ! use in a patch statement
>> ! follow with AUTOgenerate ANGLes DIHEdrals command
>>
>> !
>> ! 1HB1 1HB2
>> ! \ /
>> ! --1CB
>> ! \
>> ! 1CG==OD1
>> ! \
>> ! 2NZ- 2HZ1
>> ! |
>> ! 2HE2 - 2CE- 2HE1
>> ! |
>>
>>
>> DELETE ATOM 1OD2
>> DELETE ATOM 2HZ2
>> DELETE ATOM 2HZ3
>> GROUP
>> ATOM CB CT2A -0.27
>> ATOM HB1 HA2 0.10
>> ATOM HB2 HA2 0.10
>> ATOM CG CC 0.63
>> ATOM OD1 OC -0.75
>> ATOM CE CT2 0.22
>> ATOM HE1 HA2 0.06
>> ATOM HE2 HA2 0.06
>> ATOM NZ NH3 -0.29
>> ATOM HZ1 HC 0.34
>>
>> BOND 1CG 2NZ
>

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:23 CST