Difference between revisions of "WormBase-Caltech Weekly Calls"
From WormBaseWiki
Jump to navigationJump to searchLine 50: | Line 50: | ||
* ubc-9 gene had old (2004) manually written description but no automated description (Wen) | * ubc-9 gene had old (2004) manually written description but no automated description (Wen) | ||
* Eventually we would want all descriptions to be automated; we can't maintain manual descriptions for all genes | * Eventually we would want all descriptions to be automated; we can't maintain manual descriptions for all genes | ||
+ | * Kimberly and Ranjana will review older descriptions to see what else can be automated |
Revision as of 17:23, 18 February 2016
Contents
Previous Years
2016 Meetings
February 11, 2016
Postgres data not in ACEDB
- Should we this data get into ACEDB to be converted to Datomic (via the existing pipeline)? Maybe, but only temporarily
- We will need to learn to model in Datomic to start building curation tools that run off of Datomic
- Juancarlos has access to Colonnade & trAceView (table maker tool and web interface to view Datomic data like in ACEDB)
- For long term, we will want to build a curation tool directly on top of Datomic; will need to model currently un-dumped fields in Datomic
- Juancarlos has a few action items to test reading from and writing to Datomic in a simple manner
- Paper editor may be best form to start Datomic transition, since it doesn't deal with N1-objects changing with N2-other-objects
- Bulk updates of data not edited through the OA may be another good starting point, as those pipelines will need to change Datomic directly
CITace upload
- Upload files to Wen by Tuesday Feb 16, 10am
Life stage ontology update
- To accommodate flat worms, will make "worm life stage" root node with "platyhelminth (flatworm)..." and "nematode (roundworm)..." children
Timestamps
- Data in OA's has timestamp history, but all data first imported in bulk has one 'epoch' timestamp
- Chris looking into older timestamps from older versions of WS/ACEDB
- Will be good to clarify how older timestamps will be stored in Datomic going forward (i.e. a timestamp field separate from a Datomic intrinsic timestamp?)
- Would be best to have as many original timestamps as possible
February 18, 2016
Automated descriptions
- Plan is to phase-out the manually written descriptions
- ubc-9 gene had old (2004) manually written description but no automated description (Wen)
- Eventually we would want all descriptions to be automated; we can't maintain manual descriptions for all genes
- Kimberly and Ranjana will review older descriptions to see what else can be automated