dhimmel
commented
7 years ago Let's focus this pull request just on the sample data, i.e. generating a converted data/example/3-sample-tags.tsv. Then once that is working, we can scale to all of pubtator.
We'll only focus on the Disease, Compound, and Gene types, so other annotation types will be filtered out.
This pr closes #12 . The idea behind this pull request is to incorporate hetnet ids, when extracting pubtator tags. Hetnet ids will only appear if there the MeSH ids match the id mapper provided in the repository. Assigning review to @dhimmel and @zietzm, since you both have worked on this repository before. Let me know what you guys think.