# Regressing out variables in pearson residual normalization

**URL:** <https://discourse.scverse.org/t/regressing-out-variables-in-pearson-residual-normalization/892>\
**Category:** scanpy\
**Created:** [November 3, 2022, 10:00am UTC](https://discourse.scverse.org/t/regressing-out-variables-in-pearson-residual-normalization/892 "2022-11-03T10:00:19Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![jay](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/jay/32/429_2.png) [@jay](https://discourse.scverse.org/u/jay)\
**Post date:** [November 3, 2022, 10:00am UTC](https://discourse.scverse.org/t/regressing-out-variables-in-pearson-residual-normalization/892/1 "2022-11-03T10:00:19Z")

</div>

Hello!

I was looking into normalizing my single cell data using the Pearson Residual method in the experimental API. However, previously I regressed out mitochondrial & cell cycle differences in Seurat with SCTransform (it was a paramter within SCTransform). I was wondering in the case of Pearson Residuals, is it possible to perform the regression on the normalized Pearson Residuals?

Thank you!

---

<div class="post-metadata">

**Author:** ![redvoidling](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/redvoidling/32/608_2.png) [@redvoidling](https://discourse.scverse.org/u/redvoidling)\
**Post date:** [April 7, 2023, 11:44am UTC](https://discourse.scverse.org/t/regressing-out-variables-in-pearson-residual-normalization/892/2 "2023-04-07T11:44:01Z")

</div>

I am also curious about this especially since the output of the Pearson Residual normalisation is still very easy to separate with PCA when looking at just the cell cycle genes.

---

<div class="post-metadata">

**Author:** ![yous0614](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/yous0614/32/451_2.png) [@yous0614](https://discourse.scverse.org/u/yous0614)\
**Post date:** [July 7, 2023, 3:00pm UTC](https://discourse.scverse.org/t/regressing-out-variables-in-pearson-residual-normalization/892/3 "2023-07-07T15:00:35Z")

</div>

Follow up on this, is there an actionable workflow for apply regress\_out() in Pearson residual normalization in scanpy?