Computational Alanine Scanning CHARMM36 ff

[Navneet Kumar]

Hello! 

I want to perform Computational Alanine scanning (CAS) on a protein complex (Chain A and B). I have analysed and concluded that there are around 30 residues present at the interface of two protein. I am planning to perform the CAS by mutating all these interfacial residues. I have performed MD simulation of protein complex and obtained the trajectory. I have used the CHARMM36 ff for MD simulation. 

I would like to ask what changes do i need to make in the input parameters files for CHARMM36? 

Is it possible to perform and get del G values for mutation of all 30 residues? 

What changes do I need to make in cas_intdiel value, default it is kept 1 in the sample input files. But it probably depend on what kind of residues is mutated to Ala? 

Is there any other specific changes which I need to make in order to get the CAS results. 

____________________________________________

Sample input file for Alanine scanning

This input file is meant to show only that gmx_MMPBSA works. Althought,

we tried to used the input files as recommended in the Amber manual,

some parameters have been changed to perform more expensive calculations

in a reasonable amount of time. Feel free to change the parameters 

according to what is better for your system.

&general

sys_name="Alanine_Scanning",

startframe=1,

endframe=10,

forcefields="leaprc.protein.ff14SB", 

PBRadii=4,

/

&gb

igb=8, saltcon=0.150,

/

&alanine_scanning

mutant='ALA', mutant_res='A:23', cas_intdiel=1

/

____________________________________________

Regards

[marioe911116]

Computational Alanine scanning (CAS) on gmx_MMPBSA is supported for only one amino acid at a time. In your case, if you want to save calculation time you could run the calculation for the mutant only (mutant_only=1). 'cast_intdiel = 1' does exactly what you're asking for, a different dielectric constant depending on the mutated amino acid. So, at the end will be something like this:

&general

sys_name="Alanine_Scanning_mut_#1",

startframe=1,

endframe=10,

forcefields="leaprc.protein.ff14SB", 

PBRadii=4,

/

&gb

igb=8, saltcon=0.150,

/

&alanine_scanning

mutant='ALA', mutant_res='A:23', cas_intdiel=1, mutant_only=1

/

if you use 'mutant_only=1' then you won't have the DG values for the wild-type anymore in the output files... you will need to run an additional standard GB calculation for the wild-type system, which is not a big deal considering all the time you saved during CAS calculations... I will suggest putting everything in place (I assume you will have some sort of code to run everything) and run it with few frames, analyze the output and if everything is ok, then proceed... you don't wan to realize something went wrong after spend lot of time calculating :)

hope this helps!

[Navneet Kumar]

Thank you! 

You saved my time and effort. Otherwise I would have repeatedly calculated the delG for wild type. 

I have few other queries? 

I find this while reading tutorial. 

____________________________________________

Also, The dielectric constant (intdiel) will be modified depending on the nature of the residue to be mutated as cas_intdiel=1. In this case, the residue A/23 is a Tyrosine which means intdiel = 3 will be used.

_________________________________________________

How the cas_intdiel=3 for Tyr-Ala. from where these values will be defined? 

is there anything available like table or information in some specific literature. Because the CAS deldelG will be to small ~ 2-3 kcal/mol. And variations in results due to cas_intdiel will make wrong determination of hot spot residues. 

What other changes do I need to make specifically for CHARMM ff (like ff14SB; and igb values?)? 

--
You received this message because you are subscribed to the Google Groups "gmx_MMPBSA" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gmx_mmpbsa+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gmx_mmpbsa/d214dbb5-6a78-4d60-8066-f5536e9faa9an%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[marioe911116]

Check the &alascan namelist in teh docs to see the different variables and their allowed values (https://valdes-tresanco-ms.github.io/gmx_MMPBSA/input_file/#alanine_scanning-namelist-variables)... there is detailed explanation where the intdiel values for cast_intdiel are coming from... 

Regarding the parameters for charmm, since it is mandatory to define the topology, then the forcefields variable is dismissed... 

On the other hand, it is recommended to use PB for systems charmm force field, although you could use GB if you are planning to do relative binding free energy comparisons (any error will be canceled since is the same for all the systems) (check this tutorial on how to perform calculation for charmm systems https://valdes-tresanco-ms.github.io/gmx_MMPBSA/examples/Protein_ligand_CHARMMff/)

hth! 

[Navneet Kumar]

Thanks! 

To view this discussion on the web visit https://groups.google.com/d/msgid/gmx_mmpbsa/417c6144-245a-41a7-a3c5-02474ff76914n%40googlegroups.com.