Changes between Version 60 and Version 61 of RelationsAndTransitivity


Ignore:
Timestamp:
Mar 11, 2015, 10:33:05 AM (6 years ago)
Author:
mah79
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • RelationsAndTransitivity

    v60 v61  
     1= Describing how annotation is propagated over relations in FYPO =
     2
     3Note: It isn't really correct to refer to ancestors and descendants for relations other than ''is_a'' and ''part_of''.
     4
     5== output_of ==
     6
     7Propagation: Annotations should propagate over ''output_of'' ('''A''' ''output_of'' '''B''': all genes annotated to '''A''' should have annotation to '''B''' inferred).
     8
     9Meaning: Phenotype '''A''' results from phenotype '''B'''. A mutation that causes phenotype '''A''' must also cause phenotype '''B'''.
     10
     11Search results: A query for '''A''' should retrieve both genes annotated directly to '''A''' and genes annotated to '''B'''.
     12
     13Canto: The term page for '''B''' should offer '''A''' among the suggested terms.
     14
     15=== Example: ===
     16 * Relation in FYPO: '''A''' FYPO:0000284 large and small daughter nuclei ''output_of'' '''B''' FYPO:0001270 complete but unequal mitotic sister chromatid segregation
     17  * Meaning: If large and small daughter nuclei are formed, you can infer that complete but unequal mitotic sister chromatid segregation took place.
     18  * Search results: A query for FYPO:0000284 should retrieve genes annotated directly to FYPO:0000284 and genes annotated to FYPO:0001270.
     19  * Canto: The term view for "FYPO:0001270 complete but unequal mitotic sister chromatid segregation" should offer "FYPO:0000284 large and small daughter nuclei" as a suggestion.
    120
    221
    3 = Describing How Annotation is Propagated over relations in FYPO =
     22== has_output ==
    423
    5 Note: It isn't really correct to refer to ancestors and descendants for some of these relations
     24Propagation: Annotations should propagate over ''has_output'' ('''A''' ''has_output'' '''B''': all genes annotated to '''A''' should have annotation to '''B''' inferred).
    625
    7 [[br]]
    8 [[br]]
     26Meaning: Phenotype '''A''' always leads to phenotype '''B'''. A mutation that causes phenotype '''A''' therefore also causes phenotype '''B'''.
    927
    10 = output of =
     28Search results: A query for '''B''' should retrieve both genes annotated directly to '''B''' and genes annotated to '''A'''.
    1129
    12 '''Examples:'''
     30Canto: The term page for '''B''' should offer '''A''' among the suggested terms.
    1331
    14   * '''A.''' FYPO:0000284 large and small daughter nuclei  '''output_of B'''  FYPO:0001270 complete but unequal mitotic sister chromatid segregation
     32=== Examples: ===
     33 * Relation in FYPO: '''A''' FYPO:0002797 decreased protein degradation ''has_output'' '''B''' FYPO:0000836 increased protein level
     34  * Meaning: If protein degradation is decreased, the level of some protein will be increased.
     35  * Search results: A query for FYPO:0000836 should retrieve genes annotated directly to FYPO:0000836 and genes annotated to FYPO:0002797.
     36  * Canto: The term view for "FYPO:0000836 increased protein level" should offer "FYPO:0002797 decreased protein degradation" as a suggestion.
    1537
    16 Annotations should propagate over '''output_of''' ('''A output_of B''' : all of A should annotated to B)
    17   * i.e. In the PomBase Query builder all '''FYPO:0000284 large and small daughter nuclei''' should be annotated to '''FYPO:0001270 complete but unequal mitotic sister chromatid segregation'''
    18 
    19 In Canto we should see the '''output_of''' related terms as descendants
    20   * i.e.    '''FYPO:0000284 large and small daughter nuclei''' should be visible as a descendent of '''FYPO:0001270 complete but unequal mitotic sister chromatid segregation'''
    21 
    22 [[br]]
    23 [[br]]
     38 * Relation in FYPO: '''A''' FYPO:0001009 abolished actomyosin contractile ring assembly ''has_output'' '''B''' FYPO:0001008 actomyosin contractile ring absent
     39  * Meaning: If contractile ring assembly is abolished, no contractile ring will be present.
     40  * Search results: A query for FYPO:0001009 should retrieve genes annotated directly to FYPO:0001009 and genes annotated to FYPO:0001008
     41  * Canto: The term view for "FYPO:0001008 actomyosin contractile ring absent" should offer "FYPO:0001009 abolished actomyosin contractile ring assembly" as a suggestion.
    2442
    2543
    26 = has_output =
     44== has_part ==
    2745
    28 '''Examples:'''
    29   * '''A.''' FYPO:0002797 decreased protein degradation '''has_output B''' FYPO:0000836 increased protein level
    30   * '''A.''' FYPO:0001009 abolished actomyosin contractile ring assembly  '''has_output B''' FYPO:0001008 actomyosin contractile ring absent
    31 Annotations should propagate over '''has_output''' ('''A has_output B''' : all of A should annotated to B), i.e in the PomBase Query builder:
    32   * All '''FYPO:0002797 decreased protein degradation''' should be annotated to '''FYPO:0000836 increased protein level'''
    33   * All ''' FYPO:0001009 abolished actomyosin contractile ring assembly' should be annotated to  ''' FYPO:0001008 actomyosin contractile ring absent'''
     46Propagation: Annotations should propagate over ''has_part'' ('''A''' ''has_part'' '''B''': all genes annotated to '''A''' should have annotation to '''B''' inferred).
    3447
    35 In Canto we should see the '''has_output''' related terms as ancestors
    36   * i.e.    '''FYPO:0002797 decreased protein degradation''' should be visible as a descendent  '''FYPO:0000836 increased protein level'''
    37   * i.e.    '''FYPO:abolished actomyosin contractile ring assembly''' should be visible as a descendent  of '''FYPO:0001008 actomyosin contractile ring absent'''
    38 [[br]]
    39 [[br]]
     48Meaning: A mutation that causes phenotype '''A''' causes a phenotype with feature '''B'''.
    4049
    41 = results_in_formation_of =
     50Search results:  A query for '''B''' should retrieve both genes annotated directly to '''B''' and genes annotated to '''A'''.
    4251
    43 Annotations should propagate over results_in_formation_of
    44 results_in_formation_of A all of B should annotated to A
     52Canto: For abnormal phenotypes, the term page for '''B''' should offer '''A''' among the suggested terms. For normal phenotypes, the desired behavior is different, because a large-scale process or structure is normal only if all of its parts are normal. We therefore want to suppress the display of terms linked by has_part in the "normal phenotype" branch of FYPO (although this is not yet implemented as of this draft, 2015-03-11). See https://sourceforge.net/p/pombase/curation-tool/770/
    4553
    46 [[br]]
    47 [[br]]
     54'''Note:''' This should ONLY be done for FYPO; GO annotations should (continue to) not be propagated over ''has_part'' in either direction.
    4855
    49 = has_part =
     56=== Examples:  ===
     57 * Relation in FYPO: '''A''' FYPO:0001490 inviable elongated vegetative cell ''has_part'' '''B''' FYPO:0000049 inviable cell
     58  * Meaning: A vegetative cell that is inviable and elongated is elongated.
     59  * Search results: A query for FYPO:0000049 should retrieve genes annotated directly to FYPO:0000049 (if any, but this term should not be used directly) and genes annotated to FYPO:0001490
     60  * Canto: The term view for "FYPO:0000049 inviable cell" should offer "FYPO:0001490 inviable elongated vegetative cell" as a suggestion.
     61  * '''Note:''' As of this document draft (2015-03-11), there are asserted is_a links in FYPO that duplicate the has_part links for this and many similar terms. FYPO:0001490 is therefore not a good example for testing !PomBase search or Canto display behavior at present.
    5062
    51 '''Examples:'''
    52   * '''A.''' FYPO:0001490  inviable elongated vegetative cell   '''has_part B''' FYPO:0000049 inviable cell   (NOTE I AM NOT SURE IF THIS IS A GOOD EG AS I THINK THERE ARE ALO IS_A LIKS HERE?)
    53   * '''A.''' FYPO:0001007 normal mitosis  '''has_part B''' has_part  '''B''' FYPO:0004396 normal mitotic spindle elongation
     63 * Relation in FYPO: '''A''' FYPO:0004382 meroterically attached lagging mitotic chromosomes ''has_part'' '''B''' FYPO:0000228 lagging mitotic chromosomes
     64  * Meaning: A cell that has meroterically attached lagging mitotic chromosomes has lagging mitotic chromosomes.
     65  * Search results: A query for FYPO:0000228 should retrieve genes annotated directly to FYPO:0000228 and genes annotated to FYPO:0004382
     66  * Canto: The term view for "FYPO:0000228 lagging mitotic chromosomes" should offer "FYPO:0004382 meroterically attached lagging mitotic chromosomes" as a suggestion.
    5467
    55 Annotations should propagate over '''has_part''' from the "whole" to the "part" ('''A has_part B''' : all of A should annotated to B)
    56   * i.e. In the PomBase Query builder all '''FYPO:0001490''' should be annotated to '''FYPO:0000049'''
    57  * i.e. In the PomBase Query builder all '''FYPO::0001007'' should be annotated to '''FYPO:0004396''' (THIS ONE DOESN"T WORK???)
    58 
    59 In Canto we should see the '''has_part''' related terms as ancestors
    60   * i.e.  '''FYPO:0001490  inviable elongated vegetative cell'''   should be visible as a descendent of  '''FYPO:0000049 inviable cell '''
    61 
    62 [[br]]
    63 [[br]]
    64 
    65 
    66 
    67 This should ONLY be done for FYPO; GO annotations should (continue to) not be propagated over has_part in either direction
     68 * Relation in FYPO: '''A''' FYPO:0001007 normal mitosis ''has_part'' '''B''' FYPO:0004396 normal mitotic spindle elongation
     69  * Meaning: If mitosis is normal, its parts, including spindle elongation, are normal.
     70  * Search results: A query for FYPO:0001007 should retrieve genes annotated directly to FYPO:0001007 and genes annotated to FYPO:0004396
     71  * Canto: At present (2015-03-11) term view for "FYPO:0001007 normal mitosis" should offer "FYPO:0004396 normal mitotic spindle elongation" as a suggestion. When SF !#770 is done, this should change so that the "part" terms are not suggested on the "whole" term page.