Skip to content

Scripts for identifying sites with differential error rates in mapped nanopore DRS data

Notifications You must be signed in to change notification settings

novoalab/differr_nanopore_DRS

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 

Repository files navigation

A tool for detecting modifications from Nanopore DRS errors using a low modification control

How to use:

  • Basecall & map direct RNA data (all datasets MUST be sequenced at approximately the same time with the exactly the same flowcell/kit/MinKNOW software, and basecalled and mapped in the same way with the same model, otherwise you WILL get loads of false positives. Nanopore update their pores/kits/models/software all the time, so be wary...).
  • Run differr on mapped bam files.
  • Output from differr is a bed file with the positions with significantly altered error, and a optional hdf5 file with all of the per reference base basecalls, which might be useful for further analyses.
  • Columns of bed file are:
    • chrom, start, end, name
    • score: -log10 of the FDR, rounded to nearest whole number
    • strand
    • odds ratio: the change in the ratio of matches to mismatches in the wild type compared to the mutant with low modifications. An odds ratio > 0 indicates more modifications in the WT.
    • G statistic for the comparison of pooled WT and mutant samples.
    • -log10 P value for the comparison of pooled WT and mutant samples.
    • -log10 FDR for the comparison of pooled WT and mutant samples.
    • G statistic for the homogeneity test of mutant replicates.
    • -log10 P value for the homogeneity test of mutant replicates.
    • G statistic for the homogeneity test of wild type replicates.
    • -log10 P value for the homogeneity test of wild type replicates.

About

Scripts for identifying sites with differential error rates in mapped nanopore DRS data

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%