perform ontology mapping of two public row level epidemiological COVID datasets

[realmarcin]

The two data sources are: 1. https://github.com/beoutbreakprepared/nCoV2019/tree/master/latest_data 2. https://docs.google.com/spreadsheets/d/e/2PACX-1vQU0SIALScXx8VXDX7yKNKWWPKE1YjFlWc6VTEVSN45CklWWf-uWmprQIyLtoPDA18tX9cFDr-aQ9S6/pubhtml

For 1.

• The 'symptoms' column needs to be mapped to HPO terms as possible.
• The 'chronic_disease_binary,chronic_disease' disease columns need to be mapped to MONDO/HPO as possible.

For 2.

• The 'symptom' column needs to be mapped to HPO terms as possible.

Approach:

• make a table with all the distinct values to be mapped. This will give us an idea of the scale and eyeballing and can tell us how easy it is
• run sg-annotator or oger to initial mapping
• we can automate 95%, toss some of the junk ones, and curate the rest very quickly

Once we are mapped then two different simple analysis to generate:

• global frequencies for d2p
• sets of phenotypes with a case count

[justaddcoffee]

possibly a duplicate of #115