Have you tried to include only the chain with the mutation in the mask ? This would also require you to write the second chain only once into the parmtop file.
In practice, there is really no need to duplicate more molecules than necessary.
________________________________________
From: jfliu.itcs.ecnu.edu.cn [jfliu.itcs.ecnu.edu.cn]
Sent: 03 April 2016 14:29
To: amber.ambermd.org
Subject: [AMBER] How to use tiMerge properly in pmemd TI calculation
Dear Amber users,
I want to calculate the ralative binding free energy for a ligand to
the mutated protein through the TI calculation. However, I got
something wrong when I tried to get the merged prmtop file. The
protein has two chains, in which the residue sequence are 1-99 and
100-198, respectively. The mutation is on residue 32. To get the
merged prmtop file, I used the following script:
#!/bin/sh
#
#
parmed="python $AMBERHOME/bin/parmed.py"
$parmed protein_vdw_bonded.parm7 <<_EOF
loadRestrt protein_vdw_bonded.rst7
setOverwrite True
tiMerge :1-198 :199-396 :32 :230
outparm merged_protein_vdw_bonded.parm7 merged_protein_vdw_bonded.rst7
quit
_EOF
And some of the output was:
> Merging molecules [:1-198] [:199-396] with sc mask [:32] [:230]
MoleculeWarning: Molecule atoms are not contiguous! I am attempting to fix
this,
Merging molecules :1-198 and :199-396 into the same molecule.
Use softcore mask:
timask1='.497,498,499,500,501,502,503,504,505,506,507,508,509,510,511,512',
timask2='.3136,3137,3138,3139,3140,3141,3142,3143,3144,3145,3146,3147,3148,3149,
scmask1='.497,498,499,500,501,502,503,504,505,506,507,508,509,510,511,512',
scmask2='.3136,3137,3138,3139,3140,3141,3142,3143,3144,3145,3146,3147,3148,3149,
In the above output, I found the timask2 and scmask2 are wrong atoms. And
I checked the merged pdb file, I found the mutated residue was attached at
the end of the first chain of protein, and the correct atom number should
be 1568-1586. So in the following TI run, I got the error message "Error :
Atom 497 does not have match in V1 !".
Can anyone tell me where the problem is ? By the way, if the protein has
only one chain, I can definitely get the right timask2 and scmask2.
Thanks in advance !
Jinfeng
_______________________________________________
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 Sun Apr 03 2016 - 07:00:05 PDT
