This KBase service provides a set of methods for manipulation and visualization of expression values organized as two-dimensional matrix and assigned to genome features on one dimension and to different sorts conditions on second dimension.
The Genome object test data is only needed to provide a list of feature ids that match the ids in the paired expression data files. The prepared test data is checked into the repository as .json files in specific test/data/ directories. These .json file will work for tests on as long as the Genome object does not incur backwards incompatible changes. The version of the checked in .json files is KBaseGenomes.Genome-17.0 from ci.kbase.us circa 2/10/2020/.
The tests require three different RefSeq Genomes with source .gbff files checked in to the appropriate test.data/upload directory. The genomes are: Desulfovibrio vulgaris Hildenborough, E. coli MG1655, and Rhodobacter sphaeroides. The steps to prepare new Genome .json test data are as follows:
- Load .gbff files into a narrative in target environment.
- Download the uploaded Genome object as a .json file.
- Remove all optional _ref and handle fields in the genome object: Assembly_ref, genbank_handle_ref, gff_handle_ref, taxon_ref.
- Place edited .json Genome files in the correct test/data/upload* directory.
Roman Sutormin, LBL (rsutormin@lbl.gov) Marcin Joachimiak, LBL (mjoachimiak@lbl.gov) Pavel Novichkov, LBL (psnovichkov@lbl.gov) Srividya Ramakrishnan, CSHL (srividya.ramki@gmail.com) Michael Sneddon, LBL (mwsneddon@lbl.gov)
- There is a dependency to R-script command in both service and AWE-worker scripts
- Pyhton Scikit-learn package is not utilized yet but it's in short-term plans
- Follow the standard KBase deployment and testing procedures (you have to specify test user credentials in test/test.cfg file in order to be able to run "make test")
- to start and stop the service, use the 'start_service' and 'stop_service' scripts in (the default location) /kb/deployment/services/KBaseFeatureValues