Specifications for source files

From WormBaseWiki
Revision as of 14:20, 26 March 2013 by Vanaukenk (talk | contribs)
Jump to navigationJump to search

Specifications for Textpresso for CCC Source Files

  • The source files can be simplified a bit, but will retain the key information we need for curation and search and retrieval functions for the curation form.
  • The format will continue to be a tab-delimited file containing, in order:
  1. Sentence number in source file; starting with 1 and ending with whatever total number of sentences are in the file
  2. SSC (stands for Textpresso Sentence SCore)
  3. Numerical Textpresso sentence score value
  4. PID (stands for Paper IDentifier)
  5. Database code:numerical identifier
    1. For WormBase and dictyBase, who are sending annotations to the Protein2GO tool via web services, this identifier needs to be either a PubMed ID or a doi
    2. If neither a PubMed ID nor a doi exists, then the annotation cannot be sent to Protein2GO
    3. For TAIR, who is not yet using Protein2GO, it can still be the TAIR document ID
  6. SID (stands for Textpresso Sentence IDentifier)
  7. Numerical value of the sentence number in the document in Textpresso
  8. Gene product name or synonym as identified by Textpresso search (names below are as seen on Textpresso web sites)
    1. WormBase: protein (C. elegans)
    2. dictyBase: dicty gene
    3. TAIR: gene (arabidopsis)
  9. Textpresso component category match (names below are as seen on Textpresso web sites)
    1. WormBase: CCC cellular component 2011-02-11
    2. dictyBase: CCC TAIR
    3. TAIR: CCC TAIR
  10. Textpresso sentence (marked up version)