TAIR December 2013
TAIR, The Arabidopsis Information Resource, December 2013
IN PROGRESS, NOT READY FOR PUBLIC CONSUMPTION
1. Staff working on GOC tasks
Tanya Berardini, Donghui Li
The total number of FTE working on GOC tasks is 1.4.
2. Annotation progress
Dates are 12/12 = 12/07/2012 and 12/13 = 12/02/13.
Table 1: Number of Annotations to Various GO Aspects
|Annotations||BP (12/13)||BP (12/12)||change||MF (12/13)||MF (12/12)||change||CC (12/13)||CC (12/12)||change|
|non-IEA/non-ND||57816||57626||+ 190||13649||13268||+ 381||55665||55265||+ 400|
|IEA||17665||18361||- 696||34239||34881||- 642||7200||7331||- 131|
|ND||9206||9268||- 62||5199||5217||- 18||4471||4480||- 9|
Table 2: Number of Genes Annotated to Various GO Aspects
|Genes||BP (12/13)||BP (12/12)||change||MF (12/13)||MF (12/12)||change||CC (12/13)||CC (12/12)||change|
|non-IEA/non-ND||13335||13256||+ 79||7805||7707||+ 98||23723||23690||+ 33|
|IEA||10038||10337||- 299||12602||12795||- 193||4404||4475||- 71|
|ND||9206||9268||- 62||5199||5217||- 18||4471||4480||- 9|
3. Methods and strategies for annotation
a. Literature curation: We continue to put most of our annotation effort (95%) into annotation of gene products from the literature.
b. Computational annotation strategies: Last year, we switched to using a new program for our cellular component IEA annotations. Instead of TargetP, we now use AtSubP with an ISM evidence code. The analysis was run for TAIR by the author on the last Arabidopsis genome release, TAIR10. We are working on transitioning to using UniProtKB's IEA results for all other computational predictions to replace our aged Interpro2GO mapped ones.
c. Integration of non-TAIR Arabidopsis annotations: The following types of external annotations are included in our Arabidopsis gene association file:
1. Literature-based annotations made by TAIR curators 2. Community annotations made via TAIR's TOAST annotation tool (see below) 3. GOA annotations for Arabidopsis with experimental evidence codes 4. PAINT-based Arabidopsis annotations from RefGenome group 5. Function-Process link-based annotations from GOC 6. TIGR's annotations from Arabidopsis functional annotation project
d. Priorities for annotation:
1. literature describing the characterization of previously undescribed ('novel') genes, 2. genes that do not have any GO annotations at all (none of the three aspects), 3. recent literature from high impact factor journals
e. Review of user-submitted annotations (see TOAST section below)
Donghui and Tanya review the annotations submitted via TOAST, making sure that terms were mapped correctly and that the proper evidence_with information is entered, if necessary. Sometimes, follow-up with the submitter via email is necessary.
4. Presentations and publications
Publications - GO-related but not primarily about GO
Talks - GO-related but not primarily about GO
Donghui Li, GO task at BioCreative IV. April, 2013. Biocuration 2013. Cambridge, UK.
5. Other Highlights
A. Ontology Development Contributions
- GO terms contributed by TAIR
1. GO:1990058 Label: fruit replum development 2. GO:1990059 Label: fruit valve development 3. GO:1902347 name: response to strigolactone 4. GO:1902348 name: cellular response to strigolactone 5. GO:1902446 Label: regulation of shade avoidance 6. GO:1902447 Label: negative regulation of shade avoidance 7. GO:1902448 Label: positive regulation of shade avoidance
- Other ontology development work
- continues to participate in creating cross-products for terms within and among the three GO namespaces
- continues to participate in creation, testing, and deployment of new TermGenie templates for use by the broader community
- participated in Modular Annotation prototyping and modeling
- working on aligning the Plant Ontology (PO) with GO with respect to anatomical structures used in the development branch of the biological process ontology
- participates in the rota for Sourceforge requests (6 total people in rota, one week at a time, weekly conference call)
- participates in the rota for the gatekeeper for the TermGenie requests (6 total people in rota, one week at a time)
- attends weekly GO editors conference call
- attends regular GO annotation conference calls as the TAIR representative
- continues to request and create GO terms related to Arabiodopsis and plants
B. Annotation outreach and user advocacy efforts
- PAINT annotation
Donghui Li does PAINT-based annotation.
- attends regular PAINT annotation conference calls
- participated in the PAINT workshop at Stanford (December 2012)
- contributed to the development of PAINT SOP and user guide http://gocwiki.geneontology.org/index.php/PAINT_SOP
- testing and improvement (debugging, feature request) of PAINT tool
- curated eight PANTHER families
- TOAST (TAIR Online Annotation Submission Tool)
TAIR continues to collect controlled vocabulary annotations via its online tool. TOAST
TAIR can accept annotations based on any journal article, regardless of the journal it was published in, provided that the article has a DOI or a PMID. Submitters must be registered at TAIR.
- GO help
Tanya Berardini continues to participate in manning the GO helpdesk. This involves answering the questions that come in through the gohelp email address or forwarding them to the appropriate parties for response. There are 8 GOC curators that rotate this task, one week at a time.
TAIR took on management of the JIRA system for the GO helpdesk on June 25, 2013. All gohelp questions are now handled through jira.geneontology.org. There have been several large spam batches that have come through but the google spam filter has now caught up to them (uggs).
C. Other highlights
Donghui Li continues to serve as a member of the BioCreative User Advisory Group and an organizer for the GO task for BioCreative IV (2013). This involves developing a GO task for BioCreative V (development of modules to aid GO curators in identifying articles with curatable GO information (triage) and extracting gene function terms and the associated evidence sentences in full-length articles) as well as a strategy to create a gold-standard annotated literature corpus for text mining. Currently a paper is being written to present the results for this task.