search

nf-core / crisprseq

A pipeline for the analysis of CRISPR edited data. It allows the evaluation of the quality of gene editing experiments using targeted next generation sequencing (NGS) data (`targeted`) as well as the discovery of important genes from knock-out or activation CRISPR-Cas9 screens using CRISPR pooled DNA (`screening`).
https://nf-co.re/crisprseq
MIT License
25 stars 25 forks source link

Improve documentation describing day0 label #165

Closed jeremymsimon closed 1 month ago

jeremymsimon commented 3 months ago

Description of feature

I've read the crisprseq and MAGeCK documentation a few times and the description and function of the day0_label is still a bit unclear to me, both in terms of how it should be specified (especially if there are multiple control samples/replicates) and what specifically it does.

Let's say I have 3 control samples, and 3 treated samples, should the day0_label be set to control as is specified in my contrasts? Or should it be a comma-separated list of e.g. Control_Rep1,Control_Rep2,Control_Rep3? Or is it only appropriate for a true time-course experiment?

Further, according to the MAGeCK documentation the day0 label is a required parameter, but it does not seem to be required for crisprseq to run--I get MAGeCK counts and MLE results regardless. Is it figuring out a label automatically based on the contrasts, or is this a bug?

Thanks!

LaurenceKuhl commented 1 month ago

Hi @jeremymsimon this is normally now more thoroughly described, if you still find it confusing feel free to re open this issue :)