Hi,
so you are only mutating one atom type into another? This would mean that you have no dis- or reappearing atoms. I don't see why you would then want to use softcores or a three step protocol. Softcores imply here that none of the bonded terms contribute towards the TI gradient.
In a one step transformation you would not want to set crgmask. This mask sets all atom charges of this mask to zero.
The parmtops look fine to me.
You could also have a look at FESetup (
http://www.ccpbiosim.ac.uk/flagship#fes) which I currently develop to support such calculations. You would however need to work with the current development version. Contact me off-list if this interests you.
Cheers,
Hannes.
________________________________________
From: psu4.uic.edu [psu4.uic.edu]
Sent: 14 November 2014 22:32
To: AMBER Mailing List
Subject: [AMBER] pmemd.MPI thermodynamic integration (TI) relative binding free energy set up
Dear Amber community,
http://www.ccpbiosim.ac.uk/flagship#fes
We try to calculate the relative binding free energy of ligand A and
B. The only difference between them is A:.CL1 & B:.BR1. After carefully
reading the Amber 14 manual and the pmemd.MPI paper (
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3811123/), I am still not 100%
sure if the following commands and prmtop are correct. Could any guru
kindly give some comments? Thanks!
Input File
Step1: Removing charge on A:.CL1
Step2: Changing A:.CL1 to B:.BR1
Step3: Adding charge on B:.BR1
pmemd.MPI
No group file needed
http://www.ccpbiosim.ac.uk/flagship#fes
icfe = 1,
ifsc=0,
timask1=':1',
timask2=':2,
crgmask=’ :2.CL1’,
icfe = 1,
ifsc=1,
timask1=':1',
scmask1=':1.CL1',
timask2=':2', scmask2=':2.BR1',
crgmask=’ :1.CL1| :2.BR1’
icfe = 1,
ifsc=0,
timask1=':1',
timask2=':2,
crgmask=’ :1.BR1’,
prmtop
2 compound A + solvents
1 compound A + 1 compound B + solvents
2 compound B + solvents
One step transformation
Input File
Step1: Transform charge and vdw
pmemd.MPI
No group file needed
icfe = 1,
ifsc = 1,
timask1=':1', timask2=':2', scmask1=':1.CL1',
scmask2=':2.BR1',
crgmask=':1.CL1|:2.BR1',
prmtop
1 compound A + 1 compound B + solvents
Cheers,
Henry
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
--
Scanned by iCritical.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Nov 15 2014 - 01:00:02 PST
