Last week I was advised that my error messages from MMGBSA ...'cound not predict number of frames'... probably meant that something went wrong in parsing the overall trajectory into its constituents & I should look at the mdcrd files to verify that. I did & that's right, the molecules are messed up.
When I originally prepared the inputs, I did not write receptor, ligand, complex w/out water & I had been doing those retroactively for the other systems in the series I'm currently studying - a set of several ligands bound to the same protein - and those had all worked properly, leading me to suspect the Mg is the issue. But to rule that out, I re-constituted the system, saving all the files I thought I needed right from the start, and reran the MD. Those results are now in, & I'm still having the same problem, so it looks like the Mg is the issue. These are the commands I issue to tleap:
source leaprc.gaff
source leaprc.ff99SB
loadAmberPrep atp.prepi
lig = loadAmberParams atp.frcmod
pl = loadpdb mapr.pdb
p = loadpdb ns.pdb
l = loadpdb lig.pdb
saveAmberParm l mgatp.prmtop mgatp.inpcrd
saveAmberParm p ns.prmtop ns.inpcrd
saveAmberParm pl mgatpprnw.prmtop mgatpprnw.inpcrd
solvateBox pl TIP3PBOX 12.0
addions pl Na+ 0
addions pl Na+ 25 Cl- 25
saveAmberParm pl mgatppr.prmtop mgatppr.inpcrd
where atp.prepi & atp.frcmod are for ATP, from a web site referenced threads on this mailer's archive: http://www.pharmacy.manchester.ac.uk/bryce/amber#org, pl is a pdb containing the protein, plus ATP oriented as desired, as a residue, & Mg as another residue, also with the desired coordinates. Ns.pdb is the protein alone. l is a pdb file containing ATP & Mg, which I simply edited from pl. For input into MMPBSA.py, -sp is mgatppr.prmtop, -cp is mgatpprnw.prmtop, -rp is ns.prmtop & -lp is mgatp.prmtop...
I would have preferred to define MgATP as a single unit, but antechamber was unable to process the Mg.
Can someone tell me how [or if] I can calculate the free energy of MgATP binding to this protein? If it involves yet another way to set the system up & a third simulation, so be it.
Thanks!
Irene Newhouse
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Aug 09 2011 - 13:00:02 PDT
