Types of Papers in WormBase (and Postgres)

From WormBaseWiki
Revision as of 18:13, 17 August 2010 by Cgrove (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

For purposes of triage, indexing (data type identification), and fact extraction, here are the types of papers in Postgres and hence, WormBase:

  • Papers acquired from PubMed searches using keyword 'elegans'
  • Papers acquired via Author Request
    • Papers added via author request are typically papers that, for some reason, are missed in the PubMed searches. In the past, we only accepted papers on C. elegans; now, as more comparative nematode genomics is present in WormBase, we are also accepting papers about other nematodes.
  • Papers entered for writing Concise Descriptions ('for functional annotation only')
    • Some concise descriptions use non-elegans papers as references when, at the time the description was written, there may not have been much, or any, functional information available on the C. elegans gene but its ortholog in another species had been well studied. There are also papers used for earlier concise descriptions that describe disease genes, predicted localizations, or prion-like proteins.
    • The 'for functional annotation only' papers fall into two classes in Postgres: those that are marked as 'for functional annotation only' and those that aren't marked. The reason there are two classes is that we didn't implement the feature in the paper editor that allowed us to mark these papers as 'for functional annotation only' until ~2005-2006.
    • It should be noted that the 'for functional annotation only' papers, regardless of their label in Postgres, were never intended to be put through the first pass curation pipeline. Going forward, any papers marked as such in the paper editor, should be excluded from the Textpresso mark-up and automation pipelines.
    • We will need to agree on a workable strategy for dealing with the older 'for functional annotation only' papers. We can either label them as such as they come up in current searches, or try to pull them all out somehow (perhaps by getting the list of papers used for concise descriptions and then examining the number of first pass flags associated with those papers -- 'for functional annotation only' papers should not have first-pass flags) and then labeling them in bulk. The total number of these unlabled papers is not known.