AMBER Archive (2009)
Subject: Re: [AMBER] amber question
From: Vikas Sharma (vs_vikassharma_at_yahoo.co.in)
Hi Dr. Barbault Florent
Hope there wont be any problem
In your antechamber command, you are using AM1-BCC method. This is a semi-empirical QM method. I personnaly used several methods for my ligands :
- RESP charges, from HF 6-31* gaussian calculations. You need to do these calculations with several conformations and to do the average. This is certainly, the most accurate way to obtain ligand atomic charges.
- AM1-BCC, which seems to be less dependent to conformation. Generaly, I used it with only one ligand conformation. I have generally good results from this method but I used this method with ligands which not display large conformational changes.
- Gasteiger. This is not a quantum calculation. I employed this method for several ligands on the same RNA binding site. I found that it was not really good for my problem. However, the calculation is damn quick.
- With antechamber, you can also don't calculate charges and simply read the charges from input file. I used this for a ligand/ligand simulation in vaccuo. To do that, I took the charges from Sybyl calculations (marsili charges). This was good, but I will never do that for a protein/ligand interaction where you will need charges in adequation with amber protein charges.
To sum up, you have several possibilities depending of what you want to do. Now, I generally uses AM1-BCC as default method, and I have quite good results. In my mind, this is a good compromise with accuracy and cpu time.
Hope that my experience will help you.
On Sun, 26 Apr 2009 11:30:06 +0530 (IST)
NEW ADDRESS !!!
Universite Paris Diderot
Now surf faster and smarter ! Check out the new Firefox 3 - Yahoo! Edition http://downloads.yahoo.com/in/firefox/?fr=om_email_firefox