Skip to content

Latest commit

 

History

History
28 lines (22 loc) · 1.11 KB

miser.md

File metadata and controls

28 lines (22 loc) · 1.11 KB
Raw

MISER - Method of IES Spurious or Erroneous Reporting

This is an experimental module, may fail in unexpected ways.

Inputs

  • Mapping of PacBio HiFi/CCS reads to reference genome (preferably BAM, sorted and indexed)
  • Reference assembly
  • Feature table of putative IES coordinates, produced by MILRAA

MISER takes an existing set of IES predictions, produced by MILRAA, and screens it for potential mispredictions caused by paralogy, misassembly, or erroneous mappings.

For each putative IES (insertion or deletion), the set of reads mapping to that site is found, and split into two subsets: those containing the indel and those without. For each subset, the mean percent mismatch of alignments vs. the reference is taken.

  • If either subset has high (>5%) mismatch rate, "high error" is reported.
  • If the subset with indel has a significantly higher mismatch than the subset without, possible paralog is reported.
  • If the subset without indel has a significantly higher mismatch, possible misassembly is reported.
  • Otherwise the putative IES is "ok".