Workflow for analysis of matches to pig core microbiome @ 99%.
The mapping workflow takes two batches of metagenomes
(see inputs.mapping/lowest-detection.metags.txt and
inputs.mapping/rand_subset.3216.100.txt) and maps them against the
18 pangenomes in inputs.mapping/names.list. These metagenomes are
downloaded and trimmed and sketched with genome-grist. The config
files for genome grist are in inputs.grist/conf-lowest-detection.yml
and inputs.grist/conf-rand-100-core.yml, and you can run them with:
genome grist run conf-rand-100-core.yml smash_reads -j 24
For Figure 1:
snakemake --use-conda -j 8 do_mapping
All outputs will be placed in outputs.mapping/.
Notes:
- you'll need about 40 GB of RAM to map to the largest pangenome, 'Phocaeicola vulgatus', with 8 threads. So if you run with -j 24, you'll need to allocate 120 GB of RAM total.