search

stevencox / chemotext

1 stars 1 forks source link

p53 CTD Entries Not Tagged True #6

Open stevencox opened 7 years ago

stevencox commented 7 years ago

These rows find true relationships but are not tagged true by the evaluate phase:

#pubmed_id pubmed_date_unix_epoch_time pubmed_date_human_readable binary_a_term binary_b_term paragraph_distance sentence_distance word_distance flag_if_valid  time_until_verified

- I know p53 and mdm2 interact as well as p53 and atm but they are marked as false in eval.csv. So I looked it CTD and there are multiple entires for p53-mdm2 and p53-atm interactions
['26138448', '1435708800', '30-06-2015', '"mdm2"', '"p53"', '0', '1', '96', 'false', '1435708800\n']
['24457965', '1390435200', '22-01-2014', '"p53"', '"atm"', '0', '0', '151', 'false', '1390435200\n']
stevencox commented 7 years ago

@cpschmitt, the PubMedCentral full text mentions p53, not tp53. So there are at least two issues here.

  1. Should gene/protein names be normalized? If so, when?
  2. Is the synonym list (aggregating and translating vocabularies) happening?
cpschmitt commented 7 years ago

?The synonym list has been on hold for a bit, but if we can use both p53 and tp53 for now that would presumably let us test the p53 use case

From: Steven notifications@github.com Sent: Wednesday, October 5, 2016 9:57 AM To: stevencox/chemotext Cc: Charles Schmitt; Mention Subject: Re: [stevencox/chemotext] p53 CTD Entries Not Tagged True (#6)

@cpschmitthttps://github.com/cpschmitt, the PubMedCentral full text mentions p53, not tp53. So there are at least two issues here.

  1. Should gene/protein names be normalized? If so, when?
  2. Is the synonym list (aggregating and translating vocabularies) happening?

You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/stevencox/chemotext/issues/6#issuecomment-251682370, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AAi9fSALumHYQrlt5xp-3n5zHJEaqgyWks5qw6zjgaJpZM4KOzVi.