- 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.
-
Notifications
You must be signed in to change notification settings - Fork 3
bartongroup/differr_nanopore_DRS
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Scripts for identifying sites with differential error rates in mapped nanopore DRS data
Resources
Stars
Watchers
Forks
Packages 0
No packages published