>>>>> "jd" == jianzhong ding <jding at GRX4.BIO.BNL.GOV> writes:
jd> I have one final refined model of very high resolution
jd> structure. There is one residue (Gln) whose sidechain atoms NE2
jd> and OE1 should be switched in order to fulfill the neibouring
jd> atoms' hydrogen bond interaction. However X-PLOR always give me
jd> the negative result. Can somebody out there shed light on this?
Hi,
I'm not an X-plor user, but I've been working on this problem in
hydrogen bonding for a while. According to our software, 20000
HIS/GLN/ASN residues in 3500 released protein structures are most
probably given the wrong way around.... I guess none of the refinement
programs will be able to correct this, as "turning" the residue
requires pushing it out of the density and bumping into other residues
before there is any improvement. Furthermore, many people do not use
electrostatics during structure refinement: without those there is no
benefit at all for creating an extra hydrogen bond. Without the
hydrogen bond clues, assignment of N/O atoms from the X-ray map is
sometimes possible at better than approx 1.5 Angstrom resolution.
Something else is that in your case it may be absolutely obvious which
is the right and which is the wrong orientation. In 25% of all cases
this is not so clear (See hydrogen bonding paper by Baker and Hubbard,
1984), and only a sophisticated computer program might still give an
answer.
If you're interested take a look at:
http://www.sander.embl-heidelberg.de/whatcheck/
Regards,
--
=== Rob.Hooft at EMBL-Heidelberg.DEhttp://www.Sander.EMBL-Heidelberg.DE/rob/ ==
==== In need of protein modeling? http://www.Sander.EMBL-Heidelberg.DE/whatif/
Validation of protein structures? http://biotech.EMBL-Heidelberg.DE:8400/ ====
================ Long Live Linux! Free Software Rules The World! =============
