Dear All,
thanks to the advises of the AMBER mailing list, I have been able to
complete my calculations.
However, I have some other questions that I would like to pose you.
1. I notice that my output file prints out the following warning:
WARNING: INCONSISTENCIES EXIST WITHIN INTERNAL POTENTIAL
TERMS. THE VALIDITY OF THESE RESULTS ARE HIGHLY QUESTIONABLE
By searching on the web, I found on this post (
http://archive.ambermd.org/
201401/0010.html ) that this error is related to the fact that MMPBSA.py
makes sure that bonded interactions (specifically the BOND, ANGLE, DIHED,
and 1-4 nonbonded interactions) completely cancel out (i.e., in the Final
energy differences are zero).
They also say that in a specific case, one can be free to ignore this
warning. This is when tleap deals with improper torsions with 2 atoms of
the same type.In this case , the DIHED contribution to the binding free
energy is negligible (but still has at least one frame larger than the
0.001 kcal/mol cutoff that MMPBSA.py uses to issue a warning). In this
case, one can ignore the warning.
I think that this is my case, since I get:
Differences (Complex - Receptor - Ligand):
Energy Component Average Std. Dev. Std. Err. of
Mean
------------------------------------------------------------
-------------------
BOND 0.0000 0.0000
0.0000
ANGLE 0.0000 0.0001
0.0000
DIHED 0.0013 0.0112
0.0008
2. As you suggested, I have introduced the keyword entropy=1, in order to
perform entropy calculations too. I hope this is sufficient for performing
proper delta energy calculations including entropy contribution.
3. I am comparing the calculated energetics with 5 analogous systems in
which I have introduced mutations.
When I introduce only one mutation, the computed eneergetcs differ for ~50
kcal/mol. However, when introducing 5 mutations, I calculate a very high
differences in delta energy (i.e., 274.5471 kcal/mol for the wild type
system, 295.7256 kcal/mol for the system with one mutation, *21371334.3564
kcal/mol for the system with 4 mutatations*) . I obtain the same trend both
with Generalized Born and with Poisson Boltzmann, as well as including and
excluding the entropy term.
Looking at the simulations, I see that in the presence of 4 mutations, the
nucleic acid undergoes relevant conformational changes, even with
distortion of the backbone and fliping out of the bases.
*Can this be a reason of this very huge energy difference??? *Or do you
think that this result is due to some errors in the preparation of the
system? I have used the same input file and checked the topology. Moreover,
the BOND, ANGLE, DIHED contributions are zero (only DIHED = 0.0010).
I hope you have some advises for these issues
Thank you very much
Giulia
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Jul 02 2017 - 11:00:02 PDT
