2.5

add silly comma

2.4

fix prefix

2.3

New

• add fastq count
• add sam count

Both of which write barcode<tab>count to stdout

2.2.1

Fix

• fastq standardize was not outputting barcodes when --style wasn't provided-- it does now!!

2.2

fix counting logic

2.1.1

Fixes

• BAM/SAM filename validation for djinn sam ... commands

Full Changelog: 2.1...2.1.1

2.1

New

• migrate assign-mi from harpy to djinn sam
  • assigns MI and BX tags for unique molecules based on an alignment distance threshold
• migrate concat from harpy to djinn sam
  • concatenate SAM/BAM files while preserving within-sample molecule uniqueness
• djinn sam commands have a -S/--sam option to output SAM format instead of BAM
• specify where interleaved files wont work

Fixes

• fixed sam standardize incorrectly outputting only the first occurrence of a new barcode when -s style was invoked

Full Changelog: 2.0...2.1

2.0

Changes

• commands now need to prefixed with sam or fastq depending on the type of input file
• fastq commands accept one single-end or two paired-end files
• fastq commands have --threads option to using system pigz (if available) to speed up writing
• better fastq singleton filtering that accounts for the source of a barcode being one single-end or two paired-end reads
• downsample renamed sample

Full Changelog: 1.1...2.0

1.1

New

• filter-singletons to remove reads with singleton barcodes
• filter-invalid to remove invalid barcodes
• fastq file writing has been replaced with a fancy new internal cache object that can be configured with the hidden option --cache-size to help control gzip output speed
  • this also means no more writing uncompressed fastq and compressing after-- all fastq output is gzipped-on-write

1.0.1

Bugs fixed

• [hic-spoof] ignore invalid reads (they are moved as-is into output, no spoofing is done with them)
• [hic-spoof] fix first read in barcode pool not being added to said pool