Changes between Version 194 and Version 195 of MatrixProject


Ignore:
Timestamp:
Jan 11, 2013, 5:25:10 PM (9 years ago)
Author:
val_wood
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • MatrixProject

    v194 v195  
    99The time consuming part is creating the rules, but once the rules are in place there should be a low maintenance overhead.
    1010
    11  == Why ==
    12 Querying annotation overlaps is a powerful method to:
    13  * Identify annotation errors (curated and automated)
    14    * Allows rapid detection of incorrect mappings
    15    * Allows detection of "Indirect downstream effects". One process acts upstream of another and if the upstream process is "broken" there is an indirect downstream effect, these are usually annotated from mutant phenotypes. Would the term x involved in y make sense (for example splicing involved in embryonic development)? if not, probably the effect is indirect (I'm not sure if this is a good criteria to distinguish but it sounds as though it could be?). (Note: PomBase and SGD already do not annotate these as biological processes, they are captured with "phenotype" annotations)
    16      * Examples:
    17      * splicing -> cell cycle)
    18      * add more examples
    19  * Experimental errors (mainly from legacy data where newer papers have changed viewpoint)
    20  * Identify problems in the gene ontology
    21  * Distinguish between  direct and indirect effects and refine existing annotation
    22  * etc
    2311 
    2412 == Method ==