Hello again,
I have already managed to pinpoint the source of the error. Though I thought I was using a bugfixed version of amber, it turns out that the mmpbsa_py_energy script that was actually being used was the older version. Ensuring that the patched script is used gives more believable results for the PB calculations.
Kind regards
Pietro
________________________________________
From: amber-bounces.ambermd.org [amber-bounces.ambermd.org] on behalf of Aronica, Pietro [pietro.aronica07.imperial.ac.uk]
Sent: 30 September 2013 15:27
To: AMBER Mailing List
Subject: [AMBER] MMPBSA.py PB error
Hello,
I need to calculate the interaction between a ligand and a receptor in a complex, and I am using MMPBSA to do this.
My input file is as follows:
&general
keep_files=1
/
&gb
igb=5, saltcon=0,
/
&pb
istrng=0, inp=1
/
This will get the GB part running fine, but the PB part will fail citing this error:
CalcError: /home/pa07/Amber/amber/bin/mmpbsa_py_energy failed with prmtop 144-148_complex.prmtop!
PB Bomb in pb_aaradi(): No radius assigned for atom 7 CB 2C
I have found a workaround on the mailing list, which says to add radiopt=0 to the PB portion of the input file, which does get the calculation to run. However, the result from this calculation is very wrong.
To compare, here are the GB and PB results:
GB:
Differences (Complex - Receptor - Ligand):
Energy Component Average Std. Dev. Std. Err. of Mean
-------------------------------------------------------------------------------
VDWAALS -74.0270 7.7230 1.9941
EEL -110.6010 19.9802 5.1589
EGB 141.0885 14.4495 3.7308
ESURF -10.7044 0.6829 0.1763
DELTA G gas -184.6280 17.7664 4.5873
DELTA G solv 130.3841 14.3923 3.7161
DELTA TOTAL -54.2439 7.1754 1.8527
PB:
Differences (Complex - Receptor - Ligand):
Energy Component Average Std. Dev. Std. Err. of Mean
-------------------------------------------------------------------------------
VDWAALS -74.0270 7.7230 1.9941
EEL -110.6010 19.9802 5.1589
EPB 122.2952 15.5015 4.0025
ENPOLAR 1660.6672 80.8117 20.8655
EDISPER 0.0000 0.0000 0.0000
DELTA G gas -184.6280 17.7664 4.5873
DELTA G solv 1782.9625 84.6436 21.8549
DELTA TOTAL 1598.3345 80.7771 20.8566
As you can see, the PB result is not the usual order of magnitude that one would expect, whereas the GB one is more in line with what is normally seen.
This was all done with the latest, completely patched version of Amber. Repeating the same calculations with an older version (amber11) gives more or less exactly the same GB results (save for a few decimal figures changed here and there), while the PB gives a more usual number. This result is the same whether radiopt=0 or radiopt=1 is used.
PB (old version calculation)
Differences (Complex - Receptor - Ligand):
Energy Component Average Std. Dev. Std. Err. of Mean
-------------------------------------------------------------------------------
VDWAALS -74.0257 7.7225 1.9939
EEL -110.5992 19.9803 5.1589
EPB 122.9205 15.5350 4.0111
ECAVITY -7.9187 0.4359 0.1125
DELTA G gas -184.6249 17.7646 4.5868
DELTA G solv 115.0018 15.4784 3.9965
DELTA G binding = -69.6232 +/- 5.7039 1.4727
It seems rather clear to me that there is something wrong with the newer version of MMPBSA. Am I using the prmtop incorrectly? Is there a bugfix I haven't applied?
Kind regards
Pietro Aronica
_______________________________________________
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 Sep 30 2013 - 08:00:05 PDT
