Hi Jakob

I Just had a go at running the install script on a centos  x86_64 box here with 
the same command lines and it seemed to run the install script  fine, so I am 
not sure what the problem is. I have some suggestions below on things to do to 
check this problem out.

However, my installation on 2.33 died later in the install process because 
xplor nih 2.33 has some binary  incompatible changes to the way the ensemble 
code works [this is noted further down in the manual under tested versions].

As the distribution won’t solve the incompatibility problem can you let me know 
if you are running a stock version of xplor-nih-2.33 [ie the binary downloaded 
from charle’s site without recompilation]?  If it is a stock version I should 
be able to compile a compatible binary, otherwise I will need to know what OS 
(including version) you are running and which compilers you used to compile 
xplor.

To help me sort the installation problem can you do  a few checks for me

0. did you run md5sum on the distribution and did you get the string that 
starts 1bc59e6… that is listed next to the download link?
1. list me the content of you xplor-nih-2.33 directory with
                    ls /home/jtn/xplor-nih-2.33 and send the results to me [not 
the xplor nih list, so as to reduce traffic]
2. list the files in the distribution
                   find . <dist-dir>
                   where <dist-dir> is replaced with the path of the 
distribution directory and send the results to me [not the xplor nih list, so 
as to reduce traffic]
3. run the following modified install_dist.sh

Attachment: install_dist_logging.sh
Description: install_dist_logging.sh



and send me the results [not the xplor list, so as to reduce traffic]


On the question of 

Something went wrong, and it appears to me that cython was not properly installed; the cython dir is empty except for the __init__.py and .pyo files.
Should I install cython myself?

xcamshift doesn’t need cython installed unless you want to recompile the code…

regards
gary

-- 
-------------------------------------------------------------------
Dr Gary Thompson                    [Leeds Biological NMR Facility]

Astbury Centre for Structural Molecular Biology,
University of Leeds,
Leeds, LS2 9JT, West-Yorkshire, UK             Tel. +44-113-3433024
email: [email protected]                   Fax  +44-113-3431935
-------------------------------------------------------------------




On 25 Aug 2015, at 10:49, Jakob Toudahl Nielsen <[email protected]> wrote:

Dear Gary Thompson,

thanks for your efforts! I think this looks very promising, and I would very much like to use this module. However, I experience problems when trying to install it, hope you can help me.

I try to install by executing install_dist.sh:
--------------------
[jtn@fe2 xcamshift_1.0_linux_x86_64]$ /bin/sh ./install_dist.sh . /home/jtn/xplor-nih-2.33
installing xcamshift:

distribution directory: .
installation directory: /home/jtn/xplor-nih-2.33

cp: cannot stat `./test_data': No such file or directory
testing distribution...
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "./python/test/test_suite.py", line 17, in <module>
    from test.test_segment_manager import Test_segment_manager
  File "/home/jtn/xplor-nih-2.33/python/test/test_segment_manager.py", line 20, in <module>
    from utils import Atom_utils
  File "/home/jtn/xplor-nih-2.33/python/utils.py", line 21, in <module>
    from cython.fast_segment_manager import Segment_Manager
ImportError: No module named fast_segment_manager
PyInterp::command: error executing: >execfile('./python/test/test_suite.py')<
-----------------------------

Something went wrong, and it appears to me that cython was not properly installed; the cython dir is empty except for the __init__.py and .pyo files.
Should I install cython myself?

Hope you can help me,

best wishes,

Jakob



Jakob Toudahl Nielsen, post doc
Laboratory for Biomolecular NMR Spectroscopy
inSPIN, Center for Insoluble Protein Structures
Department of Chemistry, University of Aarhus and
Interdisciplinary Nanoscience Center (iNANO)
Gustav Wieds Vej 14, DK-8000 Aarhus C, Denmark
Office: 1593-227
Phone: +45 871 56654 (office) & 2993 8501 (cell)

Fra: [email protected] [[email protected]] på vegne af Gary Thompson [[email protected]]
Sendt: 17. maj 2015 21:51
Til: xplor-nih
Emne: [Xplor-nih] xcamshift 1.0 release

Dear All

I would like to announce the availability of XCamShift (XCS) v1.0, an xplor-nih  compatible version of the Camshift chemical shift forcefield. To quote from the manual 

• It is very thoroughly tested against the original camshift implementation in almost 
          and has a large and full test suite which checks that XCS produces results that are
          identical with those form camshift in almost.
• XCS uses cython to give a forcefield with native performance written in python
• XCS provides extensions against the standard camshift forcefield including the ability to carry out 
          ensemble calculations, weighting of shift terms and variable well widths to allow for variation in 
          statistical errors in measured chemical shift.
• XCS is modular and individual chemical shift terms can be re weighted during calculations and there
          is scope for whole components of the force field [e.g hydrogen bonding, aromatic ring currents etc] 
          to be modified or replaced independently.
• XCS provides simple human readable files defining the components of the force field.
• XCS is open source and is free software under the Library Gnu Public License (LGPL)


The website is at http://locsmith.github.io/xcamshift/ and has links for downloads of x86_64 builds for linux and osx are available along with the source code.

enjoy 

regards
gary

 
-- 
-------------------------------------------------------------------
Dr Gary Thompson                    [Leeds Biological NMR Facility]

Astbury Centre for Structural Molecular Biology,
University of Leeds,
Leeds, LS2 9JT, West-Yorkshire, UK             Tel. +44-113-3433024
email: [email protected]                   Fax  +44-113-3431935
-------------------------------------------------------------------

_______________________________________________
Xplor-nih mailing list
[email protected]
https://dcb.cit.nih.gov/mailman/listinfo/xplor-nih

Reply via email to