# What do the various CRISPR confounders mean?

**URL:** <https://forum.depmap.org/t/what-do-the-various-crispr-confounders-mean/2317>\
**Category:** Q&A\
**Tags:** genetic-screens\
**Created:** [February 21, 2023, 11:03pm UTC](https://forum.depmap.org/t/what-do-the-various-crispr-confounders-mean/2317 "2023-02-21T23:03:06Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![rnacancer](https://avatars.discourse-cdn.com/v4/letter/r/3da27b/32.png) [@rnacancer](https://forum.depmap.org/u/rnacancer)\
**Post date:** [February 21, 2023, 11:03pm UTC](https://forum.depmap.org/t/what-do-the-various-crispr-confounders-mean/2317/1 "2023-02-21T23:03:06Z")

</div>

I’ve noticed for several genes that the top predictors are confounders, especially for the CRISPR-Cas9 screens. It is possible to download a spreadsheet with all of the CRISPR confounders and values for each cell line. I can’t find where those confounders are defined. Some of them are self explanatory, others are less. What do the confounders mean (pasted below, especially the ones in bold)? Thanks  
CasActivity  
GrowthPatternAdherent  
GrowthPatternMixed  
GrowthPatternSuspension  
GrowthPatternUnknown  
LibraryAvana  
LibraryKY  
ScreenDoublingTime  
**ScreenFPR**  
ScreenMeanEssentialDepletion  
ScreenMeanNonessentialDepletion  
ScreenNNMD  
**ScreenROCAUC**  
ScreenSTDEssentials  
ScreenSTDNonessentials  
ScreenType2DS

---

<div class="post-metadata">

**Author:** ![pmontgom](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.depmap.org/pmontgom/32/4_2.png) [@pmontgom](https://forum.depmap.org/u/pmontgom)\
**Post date:** [February 24, 2023, 7:08pm UTC](https://forum.depmap.org/t/what-do-the-various-crispr-confounders-mean/2317/2 "2023-02-24T19:08:25Z")

</div>

I believe most of these should be described in the [README file for the depmap data release](https://depmap.org/portal/download/all/?releasename=DepMap+Public+22Q4&filename=README.txt).

For example, the two you have in bold are described as:

- ScreenROCAUC: area under the Receiver Operating Characteristic curve for essential (positive) vs nonessentials (negative) controls
- ScreenFPR: false positive rate, computed as fraction of nonessentials in the 15th percentile most depleted genes

Thanks,  
Phil