is it the same prmtop as you used for dynamics?
maybe there is a bond between the protein and DNA?
you could check the leap output, if you read a pdb file and it
did not have a TER line between the prot/dna or the dna strands,
this might have happened.
you also might try a minimization with sander, the output tells you
the atom with the highest energy gradient. this might help you locate the
bond.
Magne Olufsen wrote:
> We are currently running mmpbsa to calculate interaction energies
> between a protein (223 amino acids) and a DNA fragment. Sander appears
> to give very large eneriges for the bonded term. The trajectories were
> generated with SHAKE constraints on all bonds involving hydrogens,
> while the internal energies in mmpbsa is evaluated without SHAKE.
> Still it looks like somethings is wrong with the bonded term. Does
> anyone have a clue? Any help is gratefully appreciated!!
>
> Here is part of the output:
>
> MM
> GB
> PB
> MS
> 1
> BOND = 522344.8163 ANGLE = 2045.2765 DIHED =
> 2242.3395
> VDWAALS = -1848.8571 EEL = -16237.6431 EGB =
> -2875.2331
> 1-4 VDW = 886.8738 1-4 EEL = 10209.1460 RESTRAINT =
> 0.0000
> corrected reaction field energy: -4690.223908
> surface area = 10453.803
>
> Regards,
> Magne Olufsen
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Mon Dec 06 2004 - 14:53:00 PST
