Installation and quick start
Install the package
pip install sjcab_peak2anno_db
conda install stjudecab::sjcab_peak2anno_db
Install the default resource bundle:
sjcab-peak2anno-db install
sjcab-peak2anno-db install def
sjcab-peak2anno-db install default
Install selected components:
sjcab-peak2anno-db install genebed feature blacklists cgi
sjcab-peak2anno-db install -c genebed -c feature
The supported install components are genebed, feature, blacklists, and
cgi. The three commands above install the complete default bundle. list shows
the bundled registry, while path prints a resource path.
Download/install supported species/versions
sjcab-peak2anno-db download-genebed hg38 v31 -o annotations
sjcab-peak2anno-db download-feature mm10 vM22 -o annotations
sjcab-peak2anno-db install-feature cat def
sjcab-peak2anno-db install-feature dog 115
sjcab-peak2anno-db install-feature vicpac2
Use -j/--processes to process independent GTFs in parallel. GTF downloads are
sequential with a short randomized pause, and conversion within each worker is
single-process. The worker count is capped at the CPUs allocated to the job:
sjcab-peak2anno-db download-feature cat,dog,horse -j 3
dedup-genebed and filter-genebed support -n/--workers (default 2) to write
their independent TSS and TES outputs in parallel.
Python API
import sjcab_peak2anno_db as db
db.install_data()
db.download_and_convert_gencode_gtf("hg38", "v31", output_dir="annotations")
db.download_feature("hg38", "v31", "annotations")
Data lookup and cache behavior
Downloads first search the database {db-path}/cache, then the output directory, then the
current working directory. Download URLs are recorded in
{db-path}/download_urls.log. Match of gtf file name and species/version listed in
{db-path}/installed.tsv. GENCODE, UCSC, and Ensembl resolution details
are documented in the GeneBEDs guide.