Skip to content

Latest commit

 

History

History
29 lines (20 loc) · 1.51 KB

ROADMAP.md

File metadata and controls

29 lines (20 loc) · 1.51 KB
Raw

Roadmap

Thanks to everyone who expressed interest in working on this!

The purpose of this benchmarking project is to evaluate and compare a set of methods recently developed for using covariates to improve false discovery rate (FDR) estimation. Hopefully, at the end of this, we'll have a set of recommendations and/or a summary of relative strengths and weaknesses for each approach.

This "Roadmap" is a rough outline of how we can complete this project as a group.
If you have any suggestions/thoughts on the Roadmap, add them to issue #1!

Milestone 1: decide initial benchmarking setup

Tracking in issue(s) #2, #4.

  • survey previous simulations and data sets (#2)
  • agree on initial set of simulations and data sets (#4)

Milestone 2: perform benchmarking analyses

Tracking in issue(s) (not yet assigned).

  • run simulations and real data analyses (sign-up!)
  • review results as a group
  • determine follow-up analyses and update Roadmap

Milestone 3: write-up and wrap-up

Tracking in issue(s) (not yet assigned).

  • agree on format (Rmd?) and target venue of write-up
  • complete first draft and compile analyses (sign-up!)
  • get full group approval on draft
  • submit! 🎉