-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello Hongyan--

> It seems that I cannot find any tutorial for this software and the manual is
> indeed very difficult to understand. 

There's a tutorial here:
http://nmr.cit.nih.gov/xplor-nih/doc/current/python/tut.pdf

> 
> I wonder what is the difference between refine.py and anneal.py (both said to
> refine structures using NOE, RDC, J-coulpling). 

refine.py refines a rough model structure, while anneal.py determines
structure from a random extended model. If you have a starting
structure, you should use refine.py.

>  I have tried both sample
> anneal.py and my own anneal.py got warning message : TorsionTopology Warning:
> Tensor containing AXIS 500 and ANI OO is not specified and software is still
> running. I am not sure if I need axis.psf and axis.pdb file in my dir and how
> to define residue number for tensor in rdc.tbl. (I used 999 for the tensor as
> that used in Aria, and didn't define tensor.psf, but the xplor-NIH still runs
> without problems, however I am not sure if the results are reliable.)

you shouldn't get that message, and you should not usually specify any
extra axis coordinates or psf info.

> 
> In terms of determination of Da and Rh, which script should be used?
> Grid_search.inp or calTensor.py? What is the difference between
> these. 

use calcTensor unless you want to wait longer. Both scripts determine
the best tensor parameters given a set of input structures, but
calcTensor uses singular value decomposition (linear algebra).

> I have refined my structure using RDC in Aria with fixed Da, Rh and I got a
> better convergences, however, Aria can not float Da Rh, that is why I am
> switching xploy-NIH. But my initial run (anneal.py) didnot give me a better
> converged structures, is something  wrong somewhere?

It seems so. I would also use refine.py instead.

best regards--
Charles
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iD8DBQFF8W7cPK2zrJwS/lYRAvqtAJ9JjUdabhpulNFjziGae9eVa5/bIwCfbQc/
WprUu81OAlQHmmZGx7ANwXw=
=0PgF
-----END PGP SIGNATURE-----

Reply via email to