optimize preparation of promoter annotation object

GoekeLab / proActiv

Estimation of Promoter Activity from RNA-Seq data

https://goekelab.github.io/proActiv/

Other

48 stars 14 forks source link

Closed jleechung closed 4 years ago

jleechung commented 4 years ago

improve runtime for preparing promoter annotation from a transcript database (currently takes ~1 hr on Windows OS without parallelization)

jleechung commented 4 years ago

promoter annotation helper code optimized at two bottlenecks:
- replace lapply in calculating reduced exon ranges with dplyr pipe (annotation-data-helper.R, getReducedExonRanges)
- replace lapply in calculating intron ranks with dplyr pipe (annotation-data-helper.R, getIntronRanks)
restructure promoter annotation code to remove duplicate calls