I've installed Modeller9.10 on my Centos 62. linux station. I had an earlier version working under Centos 5.x. Now I find that if I set an alias I call mod9v10 to point to modeller9.10_x86_64-intel8 in analogy to the alias I successfully set to its analog previously, I get the messages about not finding libraries described on the modeller web pages. Furthermore, I get an error indicating that my .ali file can't be found & I've triple-checked to make sure I've entered the name correctly.
There is no modpy.sh in the directory to which the web pages point. Instead, there is a file modpy.sh.in. Regardless of whether I use the copy of this script in its installed location, or whether I copy it to my cwd, it cannot find the equivalent of foo.py unless I give the full path to foo.py. When I specify the full path, I get error messages such as "from: command not found" & a crash, indicative of a massive failure to access python.
If I try to use the script mod9.10, whether in its original location or a copy in cwd, it fails to find the ali file, whether or not I enter its full path.
The inputs I am using are all derived from successful modeller runs, albeit an earlier version running under an earlier Centos, so I did not expect these results. But what really makes me want to tear my hair out is this: a previous modeller manual included a short python code to extract a sequence from a PDB file. I've called it pdb2ali.py. If I type the command: mod9v10 pdb2ali.py, I get the error messages about failing to find platform dep/indep files, but I have output! There is no problem finding the input pdb file in cwd.
Please help me debug this! Thanks in advance!
Irene