Hi Dian,
It should be no problem analysing R1, R2, and NOE data at two fields
for an unassigned system. However you must be acutely aware of the
two artifacts of using an incorrect diffusion tensor in a model-free
analysis. No protein on the planet actually diffuses as a sphere
(isotropic diffusion), even if it looks roughly spherical and the
inertia tensor has 3 identical eigenvalues - a variable water shell
governed by an uneven surface charge distribution will make sure that
your diffusion is never spherical, as the water shell is part of your
diffusion tensor. The two artifacts from under or over estimating the
diffusion tensor have been described by the Schurr 1994 and Tjandra
1995 papers. I have reviewed this in full detail, with all relevant
references, in my paper at http://dx.doi.org/10.1039/b702202f. You
really must know this very well to prevent yourself of falling into
the trap of making conclusions about interesting, but artificial,
motions.
To perform such an analysis in relax, you will need some
pseudo-sequence. I'm guessing that sequence information is available
for your protein, but you just don't have an assignment yet. If you
have plans on attempting to assign this thing, then you should already
have a pseudo-sequence set up in your peak lists. This sequence can
be used directly in relax. From the relax documentation for the
spectrum.read_intensities user function (e.g.
http://www.nmr-relax.com/manual/spectrum_read_intensities.html), the
format for Sparky 'lt' peak lists is detailed. The NMRViewJ files
should just be read. As for examples, you can see a whole series of
peak lists in the test_suite/shared_data/peak_lists/ relax directory.
If you have the peak list files formatted as the relax documentation
states, have input a pseudo-sequence, but relax will still not read
the files, then this should be considered a bug. In that case, would
you be able to create a bug report for the problem at
https://gna.org/bugs/?func=additem&group=relax ? It would be best to
describe the problem in as much detail as possible - i.e. including
the script or a description of the steps you took in using the GUI.
Attaching one of the Sparky peak lists containing only one or two
residues would be appreciated, as this would allow me to reproduce the
problem.
Cheers,
Edward
On 26 June 2012 01:13, Dian Xu <dianxu@xxxxxxx> wrote:
Hi guys,
I just found this great software today but have troubles using it. What I am
dealing with is a huge protein whose structure is unknown and no sequence
information available. I have got the 15N T1,T2 and NOE data on two fields
and only want to extract the order parameter and correlation time from the
relaxation data. I am wondering whether I can do this without a sequence
file or maybe I can create a pseudo one.
Also, could someone send me some sample input files of the relaxation data
so I can modify with my own, I am not sure about the required format. I have
been tried nmrviewj and sparky to create peak list files and they do have
the data recorded in the same way mentioned in the description, however,
none of them works.
Thanks you very much!
Dian
_______________________________________________
relax (http://www.nmr-relax.com)
This is the relax-users mailing list
relax-users@xxxxxxx
To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-users
Re: No protein sequence