nielinfante / rnaseq_snakemake Goto Github PK

with conda? Running an R setup script to add libraries?

When specifying the reference in the config, it needs to be just the base, and a .fa is added in the snakefile. Lets get rid of this, but need to also change how the reference is called in make ID, salmon index, and salmon quant.

If, for example, up genes produce no results, but down genes do, get the figure for down instead of up.

Try snakemake --report

Do custom MA plot, with up genes colored, down genes colored, non-sig genes in grey and low alpha.

p-value size?
mouse-over names? Names of ten best?

Currently select only one biotype per gene. Some genes have multiple biotypes.

Solution: concatenate biotypes? pick best one?

Add option in config to output bam file from salmon. Need new conditional rule to convert output sam into bam.

When building index, make sure to include decoys

Add list of leading genes to running score slide.

• also, add a link to an explanation in of pathway mdsig.

Highlight just leading genes in pathview plot.

catch the error is vsd works on data too small.

Try making TOC open/close as well.

For each experiment, allow a message line in the config, that will be added to the report. For extra difficulty, do it on the title page.

Add an experiments folder, under that is each experiment prefix. In each of those, the report, deseq, GO, etc.