On 9/28/06, Alexandar Hansen <viochemist@xxxxxxxxx> wrote:
Hello,
I would first like to apologize that this post is going to be painfully
remedial and likely nothing more than spam to most, but I am really having
trouble getting relax to work and I don't know if it's because of me, or if
I don't have things installed correctly, or something else entirely. I'm
going to go through what I'm doing step-by-step and hopefully it will shed
light on what I'm doing wrong.
Thank you very much for running these tests. The problems are not
your fault at all. The reason you are encountering so many problems
is because you are using the data from the test-suite. I have
deliberately created these data sets to be malformed and truncated so
that it would hopefully uncover bugs in relax (which is exactly what
you are encountering). In a normal analysis, especially if you have
used relax to determine the R1, R2, and NOE, you should never
encounter such malformed data!
The reason you have found so many of these bugs is that the relax
test-suite is far from being complete. If you could create a separate
bug report for every single problem, it would be much appreciated. It
would be useful to include a link to the original post which is now
archived at https://mail.gna.org/public/relax-users/2006-09/msg00031.html,
as its message identification (Message-id:
<481156b20609271056y51ed26e3geea436983e1545af@xxxxxxxxxxxxxx>). I
will then attempt to implement tests into the test-suite to catch
these bugs and then fix the bugs one at a time.
The relax manual, I find, is very good at explaining what relax is,
describing the theory, and describing functions, but is less helpful in the
practical aspects of running relax, like a tutorial. The understanding I
came to from Chapter 2 is that relax is run by scripts, so my goal is to go
through the sample scripts and 1) run the scripts, 2) read and understand
each line comparing to Chapter 10, 3) modify the scripts for real data.
Unfortunately, my problems begin with step 1.
The relax manual is a work in progress. I am slowly writing tutorials
for each type of analysis, hopefully I will finish it one day soon.
In the meantime I'm hoping that the sample scripts together with the
individual user function documentation, just as you are currently
doing, will describe everything in detail. I wouldn't recommend
reading the user function documentation in the manual sequentially, if
anything there is too much information there. Running all the sample
scripts on the malformed data would be very useful for debugging relax
though.
The way I am writing the tutorials is to walk through certain sample
scripts describing what each step does. Chapter 4 is a complete
tutorial for calculating the steady-state NOE values, Chapter 5 is a
partially complete tutorial for the exponential curve-fitting to
determine the R1 and R2, Chapter 6 on model-free analysis is miles
from being finished, and Chapter 7 on reduced spectral density mapping
doesn't exist yet. The model-free chapter could take a while, but
anyway, I don't think it will ever be complete. For example adding a
description of setting the CSA value verse selecting the CS tensor
table for RNA/DNA as you have discussed on relax-devel would by
useful.
Thanks,
Edward
Re: frustrated in michigan