Skip to content

2. Automated Pipeline Prerequisites

Breon Schmidt edited this page Jan 21, 2021 · 5 revisions

IN PROGRESS

The preferred way for running ALLSorts should be this automated version, it takes care of everything!

Three ways to prepare counts for ALLSorts

ALLSorts is a bit sensitive as to which genes are included, to the point that just omitting any can lead to a change in result. As such, preparing counts for ALLSorts is a precise business! So we automated it for you benefit. Choose one of three methods depending on your input (fastq/fasta, bam, or raw read counts).

Before you start, you will need:

Bpipe installed - Starting with FASTQ/FASTA and BAM

Install Bpipe from Github

Bpipe is a pipeline manager that will allow you to use the automated workflows in the ALLSorts Github repository. It's an easy install. Currently, ALLSorts is tested with 0.9.9.9, but you should be able to use anything above that.

STAR Aligner - Starting with FASTQ/FASTA

Install STAR from Github

The STAR aligner will generate reference genomes and will align the reads to it. It can also generate counts that are compatible with ALLSorts, nifty!

SAMTOOLS - Starting with FASTQ/FASTA

Install SAMTOOLS from Github

Samtools will let you index the Aligned files generated with STAR for fast access!

Subread / Feature Counts - Starting with BAM

[http://subread.sourceforge.net/](Install STAR from their website]

Subreads package, Feature Counts, will help quantify the number of reads that have aligned to a feature (in our case, a gene).

Set environmental variables for executables

If your environment uses the module system, no worries, you can use that. You just need to specify the modules within bpipe.config.

Otherwise, we can set environment variables for the executables for each of the tools we will be using. Feel free to add this to your profile.

export PATH=$PATH:/path/to/bpipe/bin/

export star=/path/to/star/source/STAR

export featurecounts=/path/to/subread/bin/featureCounts

Setup bpipe.config