This is to limit clutter from old analyses.

This is used to circumvent the change in behaviour in snakemake.
https://bitbucket.org/snakemake/snakemake/issues/956/placeholders-not-properly-matched-with

Rasterizing might make pdfs lighter.

Escaping underscore for tex-compatibility is now idempotent.

Also added a missing tex-compatibility escaping.

Now using `plt.gca().set_aspect('equal', adjustable='box')`.
See <https://stackoverflow.com/a/17996099/1878788>

This looks into the results of samtools stats.

Also added test config files.

lfcMLE may bring problems with low-expressed genes.

Also:
- Boxplots with labels turned 90 degrees
- do_deseq2 returns size factors

{lib}_{treat} becomes {lib}, contrats for DESeq2 come from pairs of libs
defined in the config file.

There is now a bar graph for each library, reprensenting the raw,
deduped, trimmed and size-selected counts.

Also made make_bigwig more resilient (see
https://bioinformatics.stackexchange.com/q/2516/292).

It currently fails when libraries have too many rRNA reads:
https://github.com/arq5x/bedtools2/issues/565

Also added lfc distribution and improved MA plot.

Also moved save_plot to libworkflows.

Also fixed missing pandas import in libworkflows.

There missing file names updated in the previous commit.

"satel_si" stands for siRNAs coming from satellites.
"simrep_si" stands for siRNAs coming from simple repeats.

Fixed a missing import in libworkflows.