mayrlab/mca-utrome

3' UTR cleavage site identification from the Mouse Cell Atlas

mca-utrome's Issues

Convert R scripts to snakemake

R scripts are called through shell: sections and use positional arguments. This should be changed to use script: and then directly access the snakemake object. This removes the arbitrariness of positions, but couples more strongly to names used in the Snakefile.

Allow unfiltered index construction

Enable index with f0.0 to represent allowing all cleavages site to be included. That is, any CleanUpdTSeq scores greater than or equal to zero will be included. This requires adding some defensive code to handle the case that the unlikely subset (those failing the filter) is empty.

add readme

Identify Software

The following software needs to be converted to no longer assume it is present on PATH:

• ascp
• parallel-fastq-dump
• fastqc
• pigz
• pear
• umi_tools
• cutadapt
• rseqc - this is in a separate env py27
• R
• hisat2
• samtools
• bedtools
• python
• datamash
• snakemake

Most of this software is currently installed in my user's base Conda environment. I should be able to dump it out to a set of standalone Conda environments, making sure to note the versions.

Full-length GTF

Create an option to export a full length GTF.

Convert to use Conda environment for R

R is loaded from server modules and packages are installed locally. While we do record exact versions with #3, it would be preferable that others could easily recreate via a Conda specification.

identify kallisto version

We need to explicitly provide kallisto version, rather than assuming it is on PATH.