# Statistical test used in Custom Analysis (2-class comparison)

**URL:** <https://forum.depmap.org/t/statistical-test-used-in-custom-analysis-2-class-comparison/4622>\
**Category:** Q&A\
**Created:** [May 8, 2026, 5:33pm UTC](https://forum.depmap.org/t/statistical-test-used-in-custom-analysis-2-class-comparison/4622 "2026-05-08T17:33:56Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Princy\_Gupta](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.depmap.org/princy_gupta/32/1277_2.png) [@Princy\_Gupta](https://forum.depmap.org/u/Princy_Gupta)\
**Post date:** [May 8, 2026, 5:33pm UTC](https://forum.depmap.org/t/statistical-test-used-in-custom-analysis-2-class-comparison/4622/1 "2026-05-08T17:33:56Z")

</div>

I’ve used DepMap’s Custom Analysis 2-class comparison feature to compare three data types between two groups of cell lines :

- CRISPR Chronos scores

- Gene expression

- Metabolite abundance

I have two specific questions for each of these analyses:

1. What statistical test is used to compute the p-values, is it a simple two-sample t-test?

2. What does the effect size represent , is it the difference in means between the two groups (e.g., difference in mean Chronos score, difference in mean log2 expression)?

I’m asking because I need to accurately describe the statistical methodology in a manuscript Methods section. Does the same statistical approach apply to all three data types, or does it differ depending on the data type being analyzed?