From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Tue May 20 2008 - 08:27:50 CDT
On Tue, 20 May 2008, S.K. Ghosh wrote:
SKG> Dear users,
SKG> I am planning to set up the following simulation in NAMD. I would lke to
SKG> get your suggestion on whether this is feasible to do in NAMD.
SKG> I am trying to model the dynamics of biotin-streptavidin molecule where
SKG> biotin (thiolated) would remain attached to a Gold surface of a resonator
SKG> (QCM). So there would be a couple of layers of Au atoms, on top of it would
SKG> be a single layer of S atoms, on top of it Biotin, and attached to it one
actually the structure of thiols on gold is not that trivial.
please see for example Mazzarello et al. PRL 98, 016102 (2007).
SKG> strand of Straptavidin. I would like to investigate the forces in the
SKG> biotin-streptavidin bond as well as the S-Biotin and S-Au bond. Giving all
SKG> the atoms an acceleration I would be interested to find out which bond
SKG> gives in first, calculating the binding energy.
please note that in typical classical force fields bonds are
represented by harmonic potentials, i.e. they cannot break
and are only realistic near the equilibrium distance.
you would have to use a different functional form. but even then
i doubt that a classical potential is sufficient to describe the
behavior. see for example Krueger et al., PRL 89, 186402 (2002)
and Konopka et al., JACS 126, 1203 (2004).
you would have to do a quite complicated and time consuming
QM/MM simulation, which would probably also need you to adjust
and existing QM/MM code to be able to handle a metallic surface
system. all qm/mm "bio"-capable codes i know (and i've seen quite
a few over the last years) assume a small, isolated QM system.
SKG> Is this possible to be set-up and run in NAMD?
SKG> Is giving a constant acceleration to all atoms feasible? Is it possible to
i'd say that the constant acceleration will be the least of your
problems. have you made a reality check? how fast on the atomic scale
does your system move in the experiment? what is the period of the
p.s.: just to give you an estimate... a 5 picosecond DFT trajectory
of 16 hexane thiols on a 4-layer gold slab takes about a full month
running 24/7 on one rack of a bluegene/l (2048 cpus) (=1.5 mio cpu
SKG> get the parameter files for Au-S, Biotin-S and Biotin-Streptavidin?
SKG> Your help would be precious in my work and would be much appreciated.
SKG> Thanks very much.
-- ======================================================================= Axel Kohlmeyer akohlmey_at_cmm.chem.upenn.edu http://www.cmm.upenn.edu Center for Molecular Modeling -- University of Pennsylvania Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323 tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425 ======================================================================= If you make something idiot-proof, the universe creates a better idiot.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:21:03 CST