# Kruskal-Wallis / Wilcoxon

Both Kruskal-Wallis and Wilcoxon tests are rank tests, such rank-based tests are generally advised for use with larger sample sizes. They both can only take one factor into account at a time. Kruskal-Wallis can perform on an attribute with two or more subgroups.

Wilcoxon test is a close alternative to Kruskal-wallis task, match the results of [scany Wilcoxon method](https://scanpy.readthedocs.io/en/stable/generated/scanpy.tl.rank_genes_groups.html). This test is also called "Wilcoxon Rank-Sum Test" or "Mann-Whitney U Test". When you perform comparisons on the two groups, it will filter only include the two groups first and then perform the differential analysis.

## Running the task

To invoke the Kruskal-Wallis test, select any count-based data nodes, these include:

* Gene counts
* Transcript counts
* Normalized counts

After clicking on the chosen node:

* Select *Statistics > Differential analysis* in the context-sensitive menu.
* Select *Kruskal-Wallis or Wilcoxon.*
* Select a specific factor for analysis and click the **Next** button to setup the comparisons.

<figure><img src="https://580316046-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWMxqQAMFOJtu98OBk9KN%2Fuploads%2Fgit-blob-df202358bb47abd8a2a3fe90f08ce2306447ad52%2Fimage%20(14)%20(1)%20(1).png?alt=media" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
**Note**: Wilcoxon test will filter the data to include the observations in the two comparison groups to generate p-value, while Kruskal-Wallis will use all the samples in the input data to generate p-value on the selected attribute.
{% endhint %}

* Define the comparisons by dragging and dropping each group in the *Denominator* and *Numerator* boxes and click **Add comparison**

<div align="left"><figure><img src="https://580316046-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWMxqQAMFOJtu98OBk9KN%2Fuploads%2Fgit-blob-294df0791e90fdba157dc96c410d80f3a2067f34%2Fimage%20(327).png?alt=media" alt=""><figcaption></figcaption></figure></div>

If the data has not been filtered upstream the **Low value filter** box will be checked by default. Similarly, the **Default** normalisation will be selected if the software detects that the data as not been previously normalised.

<figure><img src="https://580316046-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWMxqQAMFOJtu98OBk9KN%2Fuploads%2Fgit-blob-3932c31f9071355cd13247ed451a2c084c041b32%2FScreenshot%202025-07-30%20at%2015.17.59.png?alt=media" alt=""><figcaption></figcaption></figure>

## Advanced option

If there are tied ranks of feature expression values, the default is not use tie correction which is corresponding to the scanpy.tl.rank\_genes\_groups(tie\_correct = False).

<div align="left"><figure><img src="https://580316046-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWMxqQAMFOJtu98OBk9KN%2Fuploads%2Fgit-blob-908a287df613ea75f905057287698c77dd69c8fd%2Fimage%20(328).png?alt=media" alt=""><figcaption></figcaption></figure></div>

The results of the analysis will appear similar to other [ANOVA/LIMMA-trend/LIMMA-voom](https://help.connected.illumina.com/icm/analyses/analysis-functionality/task-menu/statistics/differential-analysis/anova-limma-trend-limma-voom). However, the column to indicate mean expression levels for each group will display the median instead for Kruskal-Wallis.