Code for generating process-centric GO-CAM models from GAFs.

Working towards creating GO-CAMS by inputting a list of annotations (either GAF or directly ontobio association objects). Right now, this takes a GO biologcal process term as input, does some heuristic gene set calculation and generates a GO-CAM ttl for the BP term’s gene set. Separating this gene set logic from the annotation-to-GO-CAM logic is another goal.

pip install -r requirements.txt

As this is coded right now for a specific use case, this can be ran simply by inputting a GO BP term, source GAF filename, and a destination filename:

python3 generate_rdf.py -t "GO:0010971" -g "gene_association.pombase" -f "filename.ttl"

With the source GAF filename argument this now frees up the library to create GO-CAM models from any set of GAF's, not just ones pertaining to S. pombe. The example GAF can be downloaded from ftp://ftp.geneontology.org/pub/go/gene-associations/.

For my purpose right now I'm running generate_pombase_model.py specifying BP term (-t), output filename (-f), and GAF input file (-g):

python3 generate_pombase_model.py -t 'GO:0031929' -f 'TOR signaling.ttl' -g 'gene_association.pombase'

You can also specify the data (-j) to use in the first step in order to speed up processing during repeated runs (~1.5 min -> 10 sec):

python3 generate_pombase_model.py -j 'tad_go_gafs.json' -t 'GO:0031929' -f 'TOR signaling.ttl' -g 'gene_association.pombase'

To dump out this data into a reusable JSON, you can run:

python3 pombase_direct_bp_annots_query.py -j 'json_outfile.json' -g 'gene_association.pombase'

With -j specifying the JSON output path.

Requires ontobio.