[WIP] Add solvent accessibility calculation

[mlgill]

Add per-residue solvent accessibility calculation. Code is here

Notes

• External Python dependencies for the library are ProDy and Pandas
• A conda library file has been provided
• Currently also DSSP through ProDy

Open questions for feedback

• Is there a preference for how to output solvent accessible surface area -- i.e. is the absolute value in Angstroms^2 sufficient? Should the relative surface area calculation be included? And should the value be thresholded to create a binary value?
• To answer the question about thresholding, the relative surface area was plotted for each amino acid here. Unfortunately, there doesn't appear to be a clear threshold value.
• Is there a preferred format for the output data -- currently this is a CSV file
• Code is in a separate folder -- should this be incorporated with code in #29 ?
• Is a conda environment file sufficient or should a Dockerfile be created?

[gtauriello]

@mlgill @Ninjani following the conversation after https://github.com/SWISS-MODEL/covid-19-Annotations-on-Structures/issues/8#issuecomment-611939925.

I suppose that this will this be included in #29 ? It probably makes sense to have it in there as I consider the surface annotations mainly useful if we can combine them from different PDB-structures covering the same UniProt-sequence.

In terms of output-format an intermediate csv-representation for the surface accessibility makes sense and will allow to easily swap this implementation with any other implementation computing the same values. Also conda environment file should be sufficient yes.

[mlgill]

@gtauriello Thanks for the feedback -- at this point, I agree that it makes sense to combine this work with #29 . I'll close this issue to keep things tidy -- please post any additional comments / feedback there.

Regarding the CSVs, I'll leave that as an intermediate output. I'll pull the new changes that @Ninjani has made and work on finalizing the annotation upload part after tending to some family issues.