GO editors agreed on C16 relations QuickGO graph of relations much less flat 32 51 Progress on Annotation extension relations David OS transferred all C16 relations Wiki pages to GitHub and created new pages ID: 574167
Download Presentation The PPT/PDF document "Progress on Annotation extension relatio..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Progress on Annotation extension relations
GO editors agreed on C16 relations QuickGO graph of relations much less flat
32
51Slide2
Progress on Annotation extension relations
David OS transferred all C16 relations Wiki pages to
GitHub and created new pagesRachael and Ruth checked all GitHub pages
Edited where formatting errors existedRaised tickets for RO file edits requiredIdentified pages without usage examplesDavid OS edited RO files and reloaded to GitHubSlide3
Example GitHub
pagehttps://github.com/geneontology/annotation_extensions/blob/master/doc/
has_direct_input.mdSlide4
Example
GitHub
page
https://github.com/geneontology/annotation_extensions/blob/master/doc/
has_direct_input.mdSlide5
Use of relations slated for deprecation
David OS created file of AE annotations
Each annotation row unrolled to give a single AE relation per line (to enable sorting)Rachael and Ruth looked at use of dependent_on and in_presence_of
(over 4000 annotations) and extracted examples of use/misuseRama organised AE annotation calls (Rama, Petra, Ruth, Rachael provided examples)for majority of cases alternative ways of capturing were agreedIn some cases it was agreed that the example was outside range of GOhttp://wiki.geneontology.org/index.php/Annotation_Conf._Call,_July_28,
_2015 http://wiki.geneontology.org/index.php/Annotation_Conf._Call,_August_11,
_2015
Slide6
Consistent use of relations
86,600 AE annotationsRuth, Rachael, Rama reviewed relation use in all annotations
Status of term
No relations
No. annotations
Mostly
c
orrect
use
20+
>55000
Slated for
deprecation
7
~5116
edits
required*
~21
?
not yet in use (
May 2015)15
*Difficult to assess all but some are obvious from just looking:
e.g. 113 MF/BPs part_of (CL or
Uberon
) should
be
occurs_in
Slide7
Edits required
...some are obvious from just lookingSlide8
Deprecated terms?
Around 5000 annotations (~6% of extended annotations)Revise where possible?What about annotations outside of scope?
Continue to display?
dependent_on
✔
in_absence_of
?
independent_of
✔
in_presence_of
?
localization_dependent_on
✔
requires_substance
✔requires_sequence_feature
✔
agreed previously
?
still to be decidedSlide9
Completion of GitHub
page examples 6
1 relationships
# relationship
GitHub
page
Usage example
Folding example
13 *
-
-
-
25
✔
-
-
23
✔
✔
✔
Some
of the ‘completed’ examples need further
discussion
*
Causally_upstream_of
no
GitHub
page, is this controversial?Slide10
Relations still to be discussed/resolved
dependent_on – decided to deprecate at last callpart_of (GO ID) –
use with GO IDs to be revisedin_absence_of in_presence_of regulates_o_occurs_in
– confusing usage, see laterPetra’s example to be modeled in NoctuaSlide11
Discussions from calls
Exists_during
-> rename to ‘during’ to avoid confusion?
Requires_substance
-> we should not be recording cofactors, these should be in the definition
of the GO term
-> can use
activated_by
for activators (needs to be allowed for AE)
Independent_of
-> term is deprecated, but can we capture this type of concept in other ways?
-> we should not make negative terms, e.g.
parkin
-independent
mitophagy
Many annotations using controversial relations, e.g.
dependent_on
could be
expressed using a different relation or GO termSlide12
Discussions from calls cont’d
In_presence_of
/
dependent_on
-> we should avoid recording experimental assay details
-> if cofactor, this should be in GO term definition
->
GitHub
https://github.com/geneontology/annotation_extensions/issues/
17
In_absence_of
(to be discussed further)
annotations:
nucleus
in_absence_of
cAMP
cytosol in_presence_of cAMP PMID:24653039“We report here that GtaC exhibits rapid nucleocytoplasmic shuttling in response to periodic cAMP waves.” p1329Slide13
label: regulates process that occurs inLocal domain: GO:0050789 : regulation of biological
processDefinition: A relation that holds between a regulatory process and the entity which the regulated biological process or molecular function occurs in.Usage:
Use this relation to link a regulatory process or function to the structure that the regulated process or function occurs in.Currently no examples available on
GitHub pageWould expect this to just relate to BPs that involve whole cell processesRegulation of apoptosisRegulation of action potentialRegulation of mitotic cell cycleSome groups using ‘occurs_in’ rather than regulates_o_occurs_in
Regulation of transcriptionRegulation of signaling
regulates_o_occurs_inSlide14
If you are describing the regulatory process, then you should use theconventional relations (e.g.
occurs_in)If you are describing the GO process or function that is the target of theregulation, then you should use the chain relation
(e.g. regulates_o_occurs_in)'
regulation of transcription, DNA dependent' occurs_in 'fibroblast'means that the regulatory process occurs in a fibroblast‘regulation of transcription, DNA dependent' regulates_o_occurs_in 'fibroblast' means that the 'transcription, DNA dependent' occurs in a
fibroblast
regulates_o_occurs_in
cont’d
MGI guidelinesSlide15
Re-annotation?
Terms slated for deprecation
>5000 annotationsInconsistently used terms? annotations
Annotations to deprecated terms (e.g. during, not_exists_during)only 50 left!