Jeff, please see below.
On 19.12.2011 23:22, Jeffrey Thalhammer wrote:
>
> On Dec 19, 2011, at 1:39 PM, Lachele Foley (Lists) wrote:
>
>> The code is getting a little old, so do let me know if you find
>> something that doesn't work. But, there is a tool here:
>>
>> http://glycam.ccrc.uga.edu/ccrc/GlycamLITE/Protein/uploadIndex.jsp?option=ff99
>
>>
> Yes, that's exactly what I had in mind. I see that it does a nice
> job of removing the hydrogens and zeroing/stripping the columns that
> Amber doesn't want. But the cysteine residues are still CYS rather
> than CYX.
You only want to have those CYS residues "cyxified" that take part in
disulfide bonds, so you have to check manually. After cyxification in
the PDB and loading the PDB file into leap, you have to covalently bind
the CYX pairs that build up disulfide bonds. My workflow is like that
(in the shell):
1) Print residue numbers of cysteins
====================================
$ cyslist 2ILK.pdb
12 62 108 114
2) Verify disulfide bonds with literature
=========================================
12 bridges to 108
62 bridges to 114
3) cyxify all 4 cysteins
========================
$ cyscyx
USAGE EXAMPLE:
cyscyx source_PDB.pdb 12 33 17 88 199 212 > changed_PDB.pdb
-> The first argument must be the source PDB file.
All of the following arguments should be integer numbers.
All CYS residues with a residue number within the given
numbers will be replaced with CYX. The resulting PDB
content is printed to stdout.
$ cyscyx $(cyslist 2ILK.pdb) > cyxified.pdb
4) load cyxified.pdb in leap, connect disulfide bonds
=====================================================
excerpt from leap.in:
[...]
bond x.12.SG x.108.SG
bond x.62.SG x.114.SG
[...]
$ tleap -f leap.in
The executables `cyslist` and `cyscyx` are simple shell scripts (quick &
dirty solutions based on awk). You find them attached. However, I am
sure that you can build cleaner solutions in your own favourite
scripting language.
Cheers,
Jan-Philip
>
> Anyway, is the source code available under an open source license?
> Is there an API or command-line interface? The web interface is
> slick, but not practical for an automated pipeline.
>
> -Jeff _______________________________________________ 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 Dec 19 2011 - 15:30:02 PST
