question about the alaninine scanning calculation

[li tao]

Dear marioe, 

      I got a very strange results when using the alanine scanning procedure.  When I do two point mutation, such as VAL186ALA, and LEU23ALA, .  The obtained binding energy of the wildtype was very different,  one is about 56,  another is 80 kcal /mol .  The experiment results was not paralled. How to solve this problem? The following is the input file .

 &general

sys_name="alanine_scanning",

verbose=2, forcefields="oldff/leaprc.ff99SBildn",leaprc.gaff"

PBRadii=4, ions_parameters=1, temperature=318

/

&gb

igb=8, saltcon=0.150,

/

&alanine_scanning

mutant='ALA', mutant_res='A:102', cas_intdiel=1,

/

 With my best regards! 

[marioe911116]

It is difficult to spot the problem with just an input file... everything seems ok in the input file but this is definitively related to the setup of the system... feel free to share with me personally the files you are using, that way is easy for us to see what's going wrong

cheers!

[marioe911116]

When you use "cas_intdiel = 1" this automatically tells gmx_MMPBSA to use custom internal dielectric constants depending on the residue being mutated... The internal dielectric is assigned 1 when the residue being mutated is non-polar (as PRO for example), 3 when the residue is polar (as SER), and 5 when is charged... that's why you get different energies every time for the wild-type when you run the calculation for two different residues that will have different dielectric constant assigned...

read "cas_intdiel" variable in https://valdes-tresanco-ms.github.io/gmx_MMPBSA/input_file/#alanine_scanning-namelist-variables for more info

cheers!