Code Reuseability (How should we approach common analyses between codebases?)

[tlfobe]

This is still a point I think we should think and discuss a little more from our Foldamers subgroup meeting on 5/15.

Currently, some of the analysis in Foldamers requires a CGModel for the specific analyses. This means that we can't do those analyses with other representations of a foldamer without first converting to a CGModel.

I don't know what the right way to go about this is. Maybe we should use the CGModel as a central data object that we plug into all these analyses.

Some examples are:

• get_helical_distributions
• fraction_native_contacts

[mrshirts]

I think there are a couple of possibilities.

1. See what data each of the analysis routines use, and write routines that extract that information from both PDB files and CGModels.

2. Write a routine that extracts the important structural information from PDB and put them into a CGModel, even if it's not a sufficient model to actually run a simulation in - it still could be a convenient place to store that information in a useful format.