OK, I have a design question about how best to approach new residue
generation for complex molecules, (re: the acetyl-CoA discussion current
on the list).
I see two ways of going about this:
1. Treat the whole structure as one molecule.
a. This has the virtue of simplicity.
b. Dihedrals between natural substructures may or may not be well
defined.
c. There is so much flexibility (due to the flatter energy surface
for the dihedrals I mentioned) that QM programs would likely be slow
to converge.
2. Treat the structure as composed of "residues" (substructures) like we
do for nucleic acid and protein components.
a. For acetyl-CoA these would be (countless introductory biochemistry
textbooks break it down this way):
acetyl group - beta-mercaptoethylamine - pantothenic acid -
3'-5-ADP
b. Each one can be treated as a residue, the QM for each likely
converges quite rapidly, and you have a family of "residues" that
make analogs of, e.g., acetyl-CoA easy to define in LEaP. The only
problematic residue is the 3'-5'-ADP because of the extremely high
negative charge. This argues for its further subdivision into
adenosine, interior pyrophosphate, and 3'-terminal phosphate.
c. LEaP has facilities to set the dihedrals as desired and thorough
FF development will tell you the energetics (i.e., probabilities) of
the dihedral values.
d. I'm almost positive the RED people would recommend this course.
They can correct me if I am wrong.
It seems to me that method 2 is better in almost every way (although not
likely so obvious to an AMBER novice.)
What say the experts?
Bud Dodson
--
M. L. Dodson
Business email: activesitedynamics-at-gmail-dot-com
Personal email: mldodson-at-comcast-dot-net
Gmail: mlesterdodson-at-gmail-dot-com
Phone: eight_three_two-five_63-386_one
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jul 14 2011 - 08:30:03 PDT
