output.md

Directories and files produced by the tail-tools pipeline

This output has grown in an organic fashion, for which I apologize.

report/

An HTML report.

Note that this contains symlinks to other files in the directory, so when copying it make sure to dereference symlinks (eg cp -L).

plotter_config.json

JSON file containing various useful pieces of information, intended for visualization software that uses the pipeline output directory.

peaks/relation-child.gff

Peaks called from poly(A) reads. Really only the 3' end of these features is meaningful, the 3' end is the polyadenylation site.

samples/<samplename>/

Per-sample files.

Of particular importance:

• alignments_filtered_sorted.bam is the BAM file to use for viewing. These are also included as a .zip file in the html report.

The samples/<samplename>-polyA directories contain BAM files limited to reads with a poly(A) tail.

raw/

Read count and tail length statistics in CSV format. These are also given as a .zip file in the html report.

These are produced for several types of feature:

• genewise - reads anywhere in annotated genes (or some number of bases downstrand, as controlled by the extensions parameter).

• peakwise - peaks called based on poly(A) reads.

• primarypeakwise - the most prominent peak in the 3' UTR of each gene. These peaks are likely to be "real", associated with gene transcription and not some decay process. primarypeakwise is useful for investigating genomic features associated with the polyadenylation stop site.

• pairwise - the two most prominent peaks in the 3' UTR of each gene.

The files contain:

• -count.csv - read counts for each feature.

• -tail-count.csv - number of reads with a poly(A) tail for each feature.

• -tail.csv - average poly(A) tail length.