Best way to represent the protein data

[jorainer]

Extracting the protein annotations in form of a data.frame and DataFrame is straight forward, the question however is what type of object could best represent the protein annotation.

The object should be something similar to a GRanges, eventually the Proteins class from the Pbase (https://github.com/ComputationalProteomicsUnit/Pbase) package?

I've got:

• (Ensembl) protein ID with sequence.
• 1:n mapping of protein ID to Uniprot ID.
• n:m mapping between protein ID and protein domain ID, which provides in addition the position of the protein domain within the protein sequence.

@lgatto any suggestions/preferences here?

[lgatto]

I would have thought Proteins would have been the best choice, but given the circular dependency, this might not be an option. But maybe something more low-level will suffice, and we can then make use of it.

Do you want a single data structure for all the points above?

What do you mean by protein domain ID? Functional domains, or transcript exons start/end sites?

[jorainer]

regarding the protein domain ID: from Ensembl I get for each protein coding transcript its translation, which is in fact a protein sequence (AA) along with its ID (the Ensembl protein_id). For each protein_id I can then fetch the Uniprot ID (which can be none, 1 or more) and I fetch all protein domains from the various sources (Pfam, prosite, Smart); these have then start and end coordinates on the AA-sequence of the protein. That's why I thought Proteins might be a good data structure, as it allows to add to each AA-sequence also features on this sequence.

In a first version I will return protein results from the database as a AAStringSet with all additional annotations in the mcols.

[lgatto]

Fantastic! Let me know when this becomes available and I will update Pbase to make use of it.