The ABINIT Users Mailing List ( CLOSED )

["Nichols A. Romero" <naromero@gmail.com>]

Jack,

QMC would be more accurate than DFT for sure. QMC is also substantially more expensive, especially for calculating
forces which you would need if you are looking for transition states. Though it might not be too bad if your molecules
are really small. The advantage of QMC is that it is so-called "embarrassingly parallel."

The major problem with DFT is that the functionals are not particularly accurate for weak interactions like those found
in biomolecular systems. One can obtain forces rather easily though. There are a number of codes (I don't know about ABINIT) that have implemented transition state search algorithms.
http://dft.sandia.gov/Socorro/about.html

Your project web page is quite interesting. Let me know if I could be of further help.

On Dec 19, 2007 8:05 AM, < jshultz@nas.edu> wrote:

Hello,

I'm looking for a method to predict transition state structures of reaction intermediates. Reason is I'm developing a method to identify catalytic interactions based on structure data. Ultimately I hope to use these methods to datamine public databases of protein and other molecular structures for new catalysts in water splitting hydrogen production and identify an efficient alternative to fossil fuels. However before I can get to that, I need the right tool to model reactions. Some tell me quantum monte carlo and others tell me DFT. Can Abinit make estimates on reaction transition state structures? If so what parameters are necessary?

You can check out my project http://hydrogenathome.org

Happy holidays!

-- 
Nichols A. Romero, Ph.D.
DoD User Productivity Enhancement and Technology Transfer (PET) Group
High Performance Technologies, Inc.
Reston, VA
443-567-8328 (C)
410-278-2692 (O)