New GO pipeline. For context, see:
- Gathers external contributed files (GAFs, GPADs, ...)
- The locations of these are stored in yaml metadata files
- Groups can provide GAF or GPAD+GPI
- Performs filtering and checking
- Uses ontobio validate for parsing and reporting
- Afterwards runs owltools checks and inferences
- Generates filtered files and derived files
- GPAD, GAF and GPI
- RDF for direct loading into blazegraph
- Snapshot: Run every day and results go to http://snapshot.geneontology.org
- Release: Run on the 1st of each month, and goes to http://release.geneontology.org with each release in it"s own directory
- Current: http://current.geneontology.org is the current, most up to date release of products which appears in releases, above.
Everything is driven by a makefile. It should be possible to run the makefile locally, on an amazon machine slaved to jenkins, etc
The pipeline/environment.sh script can be used to setup the environment before the makefile
is run. Use source to run the script:
source environment.sh
or
. environment.sh
Active. Running in parallel. Bulk of processing and reporting is done by ontobio validate.py.
To support a more flexible processing environment and to support all annotation input types, we will change the basic structure of the pipeline, outlined here: geneontology/pipeline#206
This new pipeline "kernel" will be a series of scripts/commands that perform simple steps file-by-file. This new set of scripts/commands will be placed in the "go-association-transit" directory.
See the go-association-transit readme for more.