# Minimum sample size for MelonnPan

**URL:** <https://forum.biobakery.org/t/minimum-sample-size-for-melonnpan/2094>\
**Category:** MelonnPan\
**Created:** [May 18, 2021, 7:21am UTC](https://forum.biobakery.org/t/minimum-sample-size-for-melonnpan/2094 "2021-05-18T07:21:44Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![MinJi\_Kim](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/minji_kim/32/782_2.png) [@MinJi\_Kim](https://forum.biobakery.org/u/MinJi_Kim)\
**Post date:** [May 18, 2021, 7:21am UTC](https://forum.biobakery.org/t/minimum-sample-size-for-melonnpan/2094/1 "2021-05-18T07:21:44Z")

</div>

Hello community,  
I have a question about melonnpan.

I want to run the melonnpan with mouse gut microbiome data.  
I have only 22 samples, however I realized that melonnpan is based on ML for predicting the metabolites.

Therefore, I wonder the appropriate sample size for the MelonnPan training.

Can I use 22 samples for the training?

Thanks,  
MinJi

---

<div class="post-metadata">

**Author:** ![himel.mallick](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/himel.mallick/32/1757_2.png) [@himel.mallick](https://forum.biobakery.org/u/himel.mallick)\
**Post date:** [May 18, 2021, 12:31pm UTC](https://forum.biobakery.org/t/minimum-sample-size-for-melonnpan/2094/2 "2021-05-18T12:31:22Z")

</div>

Hi @MinJi_Kim - since you have a small sample size, you can use Leave One Out Cross Validation (LOOCV) by setting `nfolds = 22` for MelonnPan training.

All the best,  
Himel

---

<div class="post-metadata">

**Author:** ![MinJi\_Kim](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/minji_kim/32/782_2.png) [@MinJi\_Kim](https://forum.biobakery.org/u/MinJi_Kim)\
**Post date:** [May 18, 2021, 2:11pm UTC](https://forum.biobakery.org/t/minimum-sample-size-for-melonnpan/2094/3 "2021-05-18T14:11:42Z")

</div>

I really appreciate it.  
I’ll try it according to your recommendation.

Best wishes,  
MinJi