splitSwish.RdThe splitSwish function splits up the y object
along genes and writes a Snakefile that can be used with
Snakemake to distribute running swish across genes.
This workflow is primarily designed for large single cell datasets,
and so the default is to not perform length correction
within the distributed jobs.
See the alevin section of the vignette for an example. See
the Snakemake documention for details on how to run and customize
a Snakefile: https://snakemake.readthedocs.io
splitSwish(y, nsplits, prefix = "swish", snakefile = NULL, overwrite = FALSE)a SummarizedExperiment
integer, how many pieces to break y into
character, the path of the RDS files to write out,
e.g. prefix="/path/to/swish" will generate swish.rds
files at this path
character, the path of a Snakemake file, e.g.
Snakefile, that should be written out. If NULL,
then no Snakefile is written out
logical, whether the snakefile and
RDS files (swish1.rds, ...) should overwrite existing files
nothing, files are written out
Compression and splitting across jobs:
Van Buren, S., Sarkar, H., Srivastava, A., Rashid, N.U., Patro, R., Love, M.I. (2020) Compression of quantification uncertainty for scRNA-seq counts. bioRxiv. https://doi.org/10.1101/2020.07.06.189639
Snakemake:
Koster, J., Rahmann, S. (2012) Snakemake - a scalable bioinformatics workflow engine. Bioinformatics. https://doi.org/10.1093/bioinformatics/bts480