Re: error in modsel.py -- December 21, 2006

Thanks Edward for your information. I am only familar with modelfree and just
started to use relax since when I run modelfree using axially symetric models
and got some problems.
I only have data in 600 MHz and should only fit m0, m1-m5 and m9 in this 
case. I
have tried to run modsel.py using these models, still error occurred as I
posted previsouly. Not sure how to fix the problem.

Best wishes,

Hongyan

Quoting Edward d'Auvergne <edward.dauvergne@xxxxxxxxx>:

Hi Hongyan,

Chris MacRaild has replicated your problem and has created a bug
report for you at https://gna.org/bugs/?8059.  I hope you don't mind
-- I have added your email address to the CC list for bug #8056 so
that you are kept informed on how the bug is progressing.

The use of models m1 to m5 is for when you have collected the R1, R2,
and NOE at a single field strength, i.e. you have 3 data points per
spin system.  Models m1 to m5 have a maximum of 3 parameters hence can
be optimised and the resultant parameter values are statistically
meaningful.  The other models, m6 to m8, have been extensively used in
the literature when relaxation data at two or more field strengths
have been collected.  For example see:

    Korzhnev et al., Prog. NMR Spectorsc., 2001, 38(3), 197-266.
    Fushman et al., JMB, 1997, 266(1), 173-194.
    Orekhov et al., JBNMR, 1999, 14(4), 345-356.
    Zhuravleva et al., JMB, 2004, 342(5), 1599-1611.

Model m6 is specifically the full Clore et al., JACS, 1990, 112(12),
4989-4991 equation using all the extended model-free equation
parameters (S2f, tf, S2s, and ts).  Model m7 and m8 are simply models
m5 and m6 with the Rex parameter added.  The last two models m0 and m9
are where there are no statistically significant internal motions on
the picosecond to nanosecond timescales (well, according to the
relaxation data).

Cheers,

Edward


On 12/19/06, Hongyan Li <hylichem@xxxxxxxxxxxx> wrote:

Dear relax-users,
I am a new user of relax, I have tried to run modelfree analysis and get

m0, m1,

to m9 models when I cheked the results it seems m6,m7,m8 with no

parameters

fitted and m9 with Rex fitted. I don't think there is m9 in modelfree. Can

we

just fit m1,m2,m3,m4,and m5 as modelfree does? and when I run modsel.py

using

sample scripts I got error and asked me to re-run with debug flag. By

typing

relax --debug modsel.py and I got error again and it looks like that:

relax> state.save(file='relax_state_20061912_124038', dir=None, force=0,
compress_type=1)
Opening the file 'relax_state_20061912_124038.bz2' for writing.
Traceback (most recent call last):
  File "modsel.py", line 20, in ?
    eliminate()
  File "/data2/usr-local/relax-1.2.9/prompt/eliminate.py", line 97, in
eliminate
    self.relax.generic.eliminate.eliminate(run=run, function=function,
args=args)
  File "/data2/usr-local/relax-1.2.9/generic_fns/eliminate.py", line 56,

in

eliminate
    for i in xrange(num_instances(self.run)):
  File "/data2/usr-local/relax-1.2.9/specific_fns/model_free.py", line

2605, in

num_instan
ces
    raise RelaxFault
RelaxFault: RelaxError: Impossible to be here, please re-run relax with

the

'--debug' flag
 and summit a bug report at https://gna.org/projects/relax/.

Does anybody know how to do modelselection?

Very much appreciate any suggestion!

Cheers!

Hongyan



_______________________________________________
relax (http://nmr-relax.com)

This is the relax-devel mailing list
relax-devel@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-devel



Dr. Hongyan Li
Department of Chemistry
The University of Hong Kong
Pokfulam Road
Hong Kong

Re: error in modsel.py

Header

Content

Related Messages