-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello H?l?ne--
> I am running calculations with a multidomain protein (4 homologous domain,
> each 70 amino acid long), and there is a pronounced flexibility between
> domain 2 and domain 3 (revealed by RDC measured in peg and phage, and
> confirmed by relaxation) whereas the bi domains 1-2 and 3-4 appears to be
> rigid together (bidomain 3-4 even crystallizes). There are no NOE between
> individual domains.
> If I use a protocole where the RDC tensors (one for bidomain 1-2 and the
> other for bidomain 3-4) are allowed to vary (amplitude and orientation), I
> obtain a "reasonable" backbone r.m.s.d. for the full protein (3 Angtrom
> for the 271 amino acids ), with a flexibility at the level of the 2-3
> junction. Every thing seems oK.
>
> When I use a protocol where the RDC tensors are allowed to vary only in
> orientation (with the amplitude fixed (either to the values optimized in
> the previous xplor calculation, or to the values given by the software of
> Moore based on likehood), I obtain a fully rigid structure (r.m.s.d = 1.51
> for the 1-271 backbone residues)!
> I apply no gyration potential at all (due the flexibility).
> Obviously, my protocol with the fixed amplitude (which I tried to adapt
> from the tutorial examples) is not working well. Can any one help me?
Are the domain 1-2 and 3-4 structures the same for both calculations?
Are these structures very similar to the starting structure?
Your system contains multiple timescales, and particularly for the fixed
Da/Rh case, sampling is poor for overall orientation. Things wil improve
if you use Cartesian coordinates in the 2-3 linker region, and
topologically break a bond in this linker. If your the linker is resids
140 to 150, you might try:
dynRigid.breakBond("(resid 144 and name c) or (resid 145 and name n)")
protocol.cartesianTopology(dynRigid,sel="resid 140:150")
It may make more sense still to solve the two bidomains separately, and
then generate the poses of the full system in a separate ``docking''
calculation.
best regards--
Charles
- --
Charles Schwieters email: Charles at Schwieters.org
www: http://schwieters.org/cds
phone: (301) 402-4914 PGP key: http://schwieters.org/cds/pgp.txt
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.9 <http://mailcrypt.sourceforge.net/>
iEYEARECAAYFAk4M0D8ACgkQPK2zrJwS/lZXKQCggCLn1xLBL95aQaBW9IwpECZ6
XSQAnAh9sq3pwlvzHixEY/6sn4n//dN5
=ixnk
-----END PGP SIGNATURE-----
