On Thu, Dec 3, 2015 at 9:35 PM, Ziheng Wang <zihengwang.sps.edu> wrote:
> Dear Amber users,
>
> I am trying to generate prmtop and inpcrd files for a protein currently in
> mol2 format. I have tried to load the mol2 directly into tleap, change the
> naming to GAFF and load into tleap, convert to PDB and load into tleap, all
> without success. Can somebody recommend an approach?
>
> The protein mol2 file is attached.
>
I would recommend against using mol2 files as input structure files for
tleap. They are more commonly used to store residue libraries that you
would get from antechamber for a small compound. That said, I was able to
process this mol2 file and get a prmtop by converting to a PDB and fixing
it.
There are many problems with your mol2 file. First, the residue names are
decorated with numbers. Here's the start of your mol2:
.<TRIPOS>ATOM
1 N 31.1800 -1.9590 93.8660 N.am 1 ALA682
-0.4157 BACKBONE|DICT|DIRECT
2 CA 32.1570 -2.9580 94.3880 C.3 1 ALA682
0.0337 BACKBONE|DICT|DIRECT
3 C 33.3100 -3.1590 93.4100 C.2 1 ALA682
0.5973 BACKBONE|DICT|DIRECT
ALA682 is not a name that will match any of the residue templates in
Amber. So you need to get rid of those numbers. Second, for some strange
reason ALL of your hydrogen atoms come at the very end of the file. That
is, all hydrogen atoms for residue 1 come after the heavy atoms for the
last residue (251). When tleap reads this, it will think there are two
different ALA residues, one with all of its hydrogen atoms missing (which
it will helpfully fill in) and one with all of its heavy atoms missing
(which it will also helpfully fill in). The end result is twice as many
residues as you wanted. The solution there is to either move all of the H
atoms to the residues where they belong *or* delete them altogether and let
LEaP fill them in. Last comment: never use gaff for proteins. The protein
force fields are far more optimized for protein residues.
I was able to create a prmtop file from this mol2. The first thing I did
was use cpptraj to convert the mol2 into a pdb:
cpptraj -p receptor.mol2 -y receptor.mol2 -x receptor.pdb
Then I modified the residue names, deleting the number at the end of each
of the residue names (and you have to shift the columns to make sure they
match up with the PDB specification after deleting the column of numbers).
Then I deleted all of the hydrogen atoms at the bottom of the file, leaving
only the backbone and sidechain heavy atoms. This PDB file worked fine in
tleap using:
source leaprc.ff14SB
pdb = loadPDB receptor.pdb
saveAmberParm pdb receptor.parm7 receptor.rst7
(Note you would need to do any other step, like adding solvent and/or ions,
that you wanted to do).
I've attached the PDB file so you can get an idea for what it should look
like.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Dec 04 2015 - 04:30:06 PST
