STRling Versions Save

Detect novel (and reference) STR expansions from short-read data

v0.1.0

4 years ago

Major features

  • Joint genotyping: Add an optional merge subcommand which combines str read information from all sample bins to produce joint bounds that are. These loci are then genotyped in all samples.
  • Assign reads to bounds based on original cluster distances
  • provide STR disease loci for hg38
  • change default: no soft-clipped reads required for reporting loci
  • count repeats in overlapping reads (not just spanning) and don't report very high depth regions
  • only count STR loci that pass some size criteria when deciding if unplaced reads should be allocated
  • update bin file to hold version, frag dist, etc.
  • update linear models
  • Force reporting of loci from input files even if they have insufficient supporting reads (useful for comparing individuals across cohorts)

Minor features

  • Suppress reporting of several messages and output of read files to debug mode (spanning in particular can be very large for joint calling)
  • change column order of -bounds.txt to make it valid bed format and add header
  • add option to manually limit window for bounds
  • read both bounds and bed files as input at call and merge stage
  • anchored lm returns NaN when sum_str_counts is 0
  • set all bam threads to 0. Addresses #32 htslib error
  • basic end to end test run in CI for a single sample and 2 joint
  • update simulation to use merge and different insert sizes

v0.0.1

4 years ago

this is a first release of strling. to install, just grab the strling binary below, chmod+x and run

to get more information in case of a problem. run with strling_debug (which will be much slower) but will help as we iron out any issues.