[AMBER] TI with soft core potentials

From: Wood Geoffrey <geoffrey.wood.epfl.ch>
Date: Wed, 27 Jan 2010 05:18:22 +0100

Dear Amber Users,

I want to start using the TI schemes in amber10 to look at differences of binding free-energies of peptide ligands. The particular peptides that I wish to study would see mutations that would change the total charge of the system and I'm uncertain how to deal with this. Using counter ions you would obviously run in to the problem of having a charged box (by delta|1e| for charged -->neut and delta|2e| for charged --> -charged) after the transformation. I understand that sander would automatically neutralize this with a background charge but I wondered if the change in the background charge would have a large effect on the resultant gibbs free energy? For instance because the charged groups would interact differently with different backgrounds and how would the background change from one lamda window to the next?

 Alternatively I thought about mutating in addition to the peptide residues one of the counter ions but in this case you may run into the problem of having the mutated counter ion (in the case of changing one charged group into an oppositely charged group) being in a spatially undesirable place, or if you are mutating from a charged group to a neutral one then you would have to 'dissappear' the counter ion, which in the implementation in amber I'm not sure you can actually do.

The final alternative is not to add counterions at all and let both systems be neutralized by a background. But again these backgrounds are different and may interact with the charged groups in a "non-cancelling manner" And again how would the background change from one lamda window to the next?

If anyone has had experience with these types of TI problems I would appreciate any advice about how to proceed.

Thanks in advance,

Dr Geoffrey Wood
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jan 26 2010 - 20:30:02 PST
Custom Search