On 30 May 1996 09:21:24 -0700 Gregory King asked:
> I'm looking for software (or an explanation of an efficient algorithm) that
> will superimpose molecule A onto molecule B with a combination of translations
> and rotations, and using a minimal rms deviation as the convergence criterion.
> There is a known one-to-one correspondence between the relevant atoms in both
> molecules (I'll be comparing peptides, and using the backbone atoms as the
> relevant ones).
Most of the major protein simulation and modeling packages should provide
a means of doing this; CHARMM does, so I suspect so does X-PLOR. I'd
speculate that AMBER or GROMOS can also do this without too much difficulty.
If any of these packages are available, that's one possibility. Also, most
homology modeling programs should handle this as well. Your choices will
depend on:
(1) available hardware; many of the more graphics-oriented tools are only
available for SGI machines; most require a Unix workstation at a minimum,
but a Pentium running Linux uith a fast clock and extra RAM would work
(2) buy vs. download and compile; this can be the same package, i.e. some
packages are freely available to not-for-profit institutions, but private
companies are obliged to pay for them. It may depend on whether cash or
time is more precious to you.
It's hard to give more specific advice without knowing your hardware and other
resources (compilers, money for software, time) and a little more about your
affiliation (biotech industry, academia, government lab, ...)
If you have a web browser try the NIH molecular modeling pages at URLs
http://molbio.info.nih.gov/modeling/text.html (Mosaic, other browsers)
http://cmm.info.nih.gov/modeling/ (Netscape 2 or better)
--
Rick Venable =====\ |=| "Eschew Obfuscation"
FDA/CBER Biophysics Lab |____/ |=|
Bethesda, MD U.S.A. | \ / |=| / Not an official statement \
rvenable at deimos.cber.nih.gov \/ |=| \ or position of the FDA. /
