Automate gene cache population, expand from 2 to 14 organisms

[eweitz]

This encodes the process of creating gene cache files, and increases the number of supported organisms by 7x.

Previously, gene caches were artisanally constructed by hand for human and mouse. I would download a GTF file from Ensembl, pipe some greps, run replace all in various ways, and do low-code spreadsheet operations. Now, those steps are fully automated (and thus also much more clearly documented!) in gene_cache.py.

You can run that pipeline like so:

cd scripts/python
python -m venv env --copies
source env/bin/activate
pip3 install -r requirements.txt
cd cache
python3 gene_cache.py --output-dir ../../../dist/data/cache/

(Add--reuse-gtf at the end of that last command if re-running, e.g. while developing, for faster iteration.)

Organisms with gene cache available:

• Homo sapiens (human)
• Mus musculus (mouse)
• Danio rerio (zebrafish)
• Gallus gallus (chicken)
• Rattus norvegicus (rat)
• Pan troglodytes (chimpanzee)
• Macaca fascicularis (crab-eating macaque / cynologus monkey)
• Macaca mulatta (c)
• Canis lupus familiaris (dog)
• Felis catus (cat)
• Equus caballus (horse)
• Bos taurus (cow)
• Sus scrofa (pig)
• Caenorhabditis elegans (worm / nematode)

Also, the cache directory has also been moved, and various small bugs have been fixed.

[coveralls]

Coverage decreased (-0.002%) to 88.909% when pulling 43381056aafdfe6a510cd1e080e9632401bf5966 on populate-gene-cache into 259e9c87ff9495139f8e61cbd377d62ae6ce5761 on master.