# Scanpy.tl.rank\_genes\_groups with wilcoxon returns same score for multiple genes

**URL:** <https://discourse.scverse.org/t/scanpy-tl-rank-genes-groups-with-wilcoxon-returns-same-score-for-multiple-genes/2384>\
**Category:** scanpy\
**Created:** [July 15, 2024, 3:58pm UTC](https://discourse.scverse.org/t/scanpy-tl-rank-genes-groups-with-wilcoxon-returns-same-score-for-multiple-genes/2384 "2024-07-15T15:58:27Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![osmanmerdan](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/osmanmerdan/32/1097_2.png) [@osmanmerdan](https://discourse.scverse.org/u/osmanmerdan)\
**Post date:** [July 15, 2024, 3:58pm UTC](https://discourse.scverse.org/t/scanpy-tl-rank-genes-groups-with-wilcoxon-returns-same-score-for-multiple-genes/2384/1 "2024-07-15T15:58:27Z")

</div>

Hi everyone,  
Just a simple question.  
I have run `scanpy.tl.rank_genes_groups` to find DEGs between two clusters.

- I have set `method="wilcoxon"`
- Using shifted log normalized values.
- I have set `tie_correct = False` (Also, I have tried it with ‘True’)
- I have set `rankby_abs = True`

When I checked the resulting dataframe using `scanpy.get.rank_genes_groups_df`, I got many genes with duplicated scores.  
I want to do GSEA with `gseapy`, using scores. But of course, I get a warning about genes with the same score. So, do you know how to deal with these genes with identical scores? Also, is it a little bit weird to get the same score? Could it be a decimal thing?

Example dataframe for some duplicated scores:

| names | scores | logfoldchanges | pvals | pvals\_adj | pct\_nz\_group |
| --- | --- | --- | --- | --- | --- |
| A530040E14Rik | 4.767420 | 21.592094 | 0.000002 | 0.000006 | 0.005073 |
| 1700048O20Rik | 4.767420 | 21.506248 | 0.000002 | 0.000006 | 0.005073 |
| Jakmip1 | 4.697415 | -0.022416 | 0.000003 | 0.000008 | 0.181096 |
| Nrcam | 4.697415 | 0.745594 | 0.000003 | 0.000008 | 0.018262 |

I appreciate any help you can provide.  
Best