generic workflow for downloading and extracting TEs from DFAM.
snakemake --use-conda --cores <n> Expects a config containing: DFAM_H5 (a uri for a DFAM famdb formatted h5 file, can be remote) DFAM_ARGS (-ad by default, to retrieve ancestor and descendant repeats) DFAM_CLASSES (see note below)
Yields
The requested repeat sequences: results/dfam/dfam_sequences.fasta
results/dfam/dfam_summary.txt
Info about lineages from fasta header:results/dfam/dfam_lineages.txt
By default all rules run locally. This is because downloads are generally faster on login nodes (at least where I am) and the other steps are just extraction from databases with minimal resource requirements.
The allowed nomenclature is poorly documented (ref1 and ref2). But there are classes of repeat that are included in a full DFAM dump that are probably not what users want, such as simple repeats.
By default this pipeline retrieves sequences with RepeatMasker classes SINE, LINE, LTR, RC, and DNA. This is nonredundant for mus musculus and yields 1237 TEs with DFAM 3.7.
Subsets of TEs (or other repeat types) can be specified by adding the following to the config.
DFAM_CLASSES: [LINE, SINE, ...etc...]