Dear Amber users
I have developed a procedure for making and breaking disulfide bridges
"on the fly" during MD with sander. There was somebody who requested
this code from me some time ago, and I've forgotten who it is. At the
time I didn't provide it because there were some bugs....they're all
fixed now.
The algorithm works as follows:
1. The topology files for all the combinations of disulfide bridge
bonds are generated in advance.
2. Usual MD are applied for a given time step (say 2 ps to a starting
configuration.
3. After this time frame the euclidean distance of each disulfide pair
is calculated and a bond made probabalistically using the following
sigmoidal function:
p(bonding) = 1(1+exp(ax + b))
where x = euclidean distance between S-S pairs, and a and b are
constants that can be varied.
Thus as SG atoms approach one another the probabality of them bonding
increases.
4. If disulfide bridges form or break then the coordinates of the
existing system are mapped to a new (and appropriate) topology file.
The procedure is simple but it has helped me sample some cross-bridged
systems more extensively than was previously possible. If anyone out
there needs the code I'd be happy to supply it, along with a brief set
of instructions.
best
Hayden Eastwood
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Apr 25 2007 - 06:07:36 PDT
