QCQA call 2017-11-29
Meeting minutes 2017-11-29
- Suggestions for documentation: Include examples of annotations to help curators
- Documentation needs to be accessible from the annotation tool (follow up on the suggestion to create an annotation type in Protege)
- Suggestions by Protein2GO can be misleading (what to do? We probably need to evaluate the suggestions to see how they are being used)
- Should look at http://build.berkeleybop.org/view/GAF/
- Get Chris /Seth to explain
- For active groups, discuss at some annotation meeting
- For inactive groups - delete them automatically ?
- Ask to get a report for each rule to see which to prioritize
- Get a report for papers with errata/retractions ??
- What error rate are we aiming for?
- The aim is 100%. This varies a lot by term, etc
- Two types of errors:
1) Incorrect annotations: “Errors” in which the general/high level branch is OK are are not considered errors. We aim to annotate in the correct branch Pragmatic aim for now is 90% (for eg when we ask for a annotation review) 2) Errors of omission: when an ‘obvious’ annotation is missing - hard to tackle Errors: often come from older annotations/older papers -> Add to guidelines to review older annotations when curating a new paper Do not manually annotate: some should be hard checks, some should just be a soft check Improve mechanism for disputes: working well Later: Implement a mechanism that GOC database can post-process annotations to flag them (or whatever mechanism) to remove them from our dataset
- Ontology changes: It would be nice to have a tool in which we *model* the changes in slims or enrichment before committing certain ontology changes (especially when we add/remove parents)
Perhaps the way would be to look at the parent and see the difference in number of annotations
Action points
- VAL to go through the Biocuration talk about the matrix and share common errors and examples where people felt they needed to keep the annotation
- We will compile a list of inconsistent annotation practices (for example xxx )
- Prioritize the development of guidelines for these
- Ask Tony to add ORCID in EBI GPAD
- GAF reports: http://build.berkeleybop.org/view/GAF/
- Get Chris /Seth to explain
- Are soft checks reported ?
- Soft checks should be associated with a comment
- For active groups, discuss at some annotation meeting
- For inactive groups - delete them automatically ?
- Ask to get a report for each rule to see which to prioritize
- Provide a list of curators expert domains: in the contacts file
SEE ALSO
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995153/
[[Category:Quality Control]