README

We provide several scripts to obtain a StructureDB. 
More information regarding the different steps you'll find in the comments
in the corresponding scripts.

Steps to obtain a structural database:

1. Get the structure/profile data: preferably you want a non redundant set of
   structures to generate a StructureDB. We suggest to cull a list of pdb entries
   from PISCES: http://dunbrack.fccc.edu/PISCES.php
   The get_data_from_smtl.py script directly reads in such a file and produces
   the input for the second step in StructureDB generation. 

2. Build an initial database: use build_structure_db.py with input generated
   in step one.

IF YOU NEVER EVER REQUIRE A STRUCTURE PROFILE, YOU'RE ALREADY DONE!
BUT BE AWARE, THE ACCORDING FREQUENCIES IN THE DATABASE WILL BE SET TO 0.0!

3. Calculate structure profiles: use create_structure_profiles.py.
   The script calculates the structure profiles for a subset of the
   sequences in the desired StructureDB and is intended for parallelization
   
4. Parallelize step 3: use submit_structure_profile_calculations.py
   This script uses the SGE submission on BC2 and you might have to 
   adapt for your system

5. Assign the structure profiles generated in step 3 and 4 to the 
   initial database generated in step 2: 
   use assign_structure_profiles.py to perform this task