STRetch v0.4.0

Major changes:

• Introduces an exome bam pipeline: allows running STRetch directly from exomes in bam format

Minor changes/bugfixes

• Add unit tests and coverage reports
• Add tests for identify_locus.py and its dependencies, and refactor these to aid testing
• Round off floating point numbers in result STR.tsv files
• identify_locus.py sorts bed inputs by default
• Fix some paths pipeline_config_template.groovy

STRetch v0.3.4

• Important fix for a bug that resulted in sample ID mixups, see #41
• Improved error message when STRetch encounters unexpectedly low sequencing coverage, see #38

STRetch v0.3.3

• Improvements to installation, especially generating bpipe configuration to aid in using resource managers
• Fig bug with references to jar files

STRetch v0.3.2

• Retains entire input filename as sample name for bam files
• Tries to sensibly extract sample name from fastq files, but will fall back on entire filename
• Add ".STRdecoy" suffix to STRetch intermediate bam files to avoid a name clash with the input bam file
• Warn when running with a single sample and no controls
• Suppress some irrelevant warnings generated by dependancies

STRetch v0.3.1

Test for jar files during installation

STRetch v0.3.0

This version uses the Bazam tool (https://github.com/ssadedin/bazam) to extract relevant reads from a bam/cram file before remapping to the STR decoy genome. Because this tool pipes directly to BWA, can be run in parallel, and handles grabbing both reads in a pair correctly, this change both decreases runtime and increases sensitivity.

Also fixes errors caused by some versions of Samtools (see #25, #26)

STRetch v0.2.0

• Reimplement estimateSTR.R in Python (now estimateSTR.py), removing R as a STRetch dependancy
• Generate more informative error messages for example for missing bam file CIGAR strings, empty or missing intermediate files, no reads aligned to STR decoys
• Stricter p value correction
• Script to convert Tandem Repeats Finder dat format to bed format, allowing custom reference genomes to be used ('TRFdat_to_bed.py')
• Updates to comply with latest version of Pandas (v0.23.x)

STRetch v0.1.0

Features

• Version used for preprint doi: https://doi.org/10.1101/159228.
• Pipelines for using STRetch with WGS data starting from fastq or mapped bams.
• Pipeline for using STRetch with exome data starting from fastq - this is a work in progress, and size estimates are unlikely to be useful.
• Install script downloads reference data and installs all dependancies except R.
• Helper scripts
• Continuous integration for installation and running test data.