MMPBSA Calculation for Transmembrane Proteins

[Thịnh Quốc]

Dear MMPBSA Support Team,

I am currently running calculations to analyze the interactions within a ligand-protein complex. I am concerned about whether, for a transmembrane protein, it is necessary to run the MMPBSA calculation with the embedded parameters, as shown in your example on the website.

&general sys_name="Prot-Memb-CHARMM", startframe=1, endframe=4, # In gmx_MMPBSA v1.5.0 we have added a new PB radii set named charmm_radii.# This radii set should be used only with systems prepared with CHARMM force fiel# Uncomment the line below to use charmm_radii set #PBRadii=7, / &pb memopt=1, emem=7.0, indi=1.0, mctrdz=37, mthick=40, poretype=1, radiopt=0, istrng=0.150, fillratio=1.25, inp=2, sasopt=0, solvopt=2, ipb=1, bcopt=10, nfocus=1, linit=1000, eneopt=1, cutfd=7.0, cutnb=99.0, maxarcdot=15000, npbverb=1, /

Alternatively, can I run the calculation as usual without the embedded parameters? I do not see any membrane-related parameters in the &gb calculation. Can I still use it for my system?

Lastly, which method would be more suitable for a protein embedded in a membrane—PB or GB?

Thanks a lot for your supports.

[marioe911116]

In theory, this is the "most" accurate way to perform a calculation with a transmembrane protein. You could decide to perform a regular GB calculation with no embedding at all, in which case any effect of the membrane won't be accounted for. For example, this would be more suitable for cases where you want to compare different ligands against the same protein. Ultimately, it depends on you.