# Using Melonnpan wiht HUMAnN4

**URL:** <https://forum.biobakery.org/t/using-melonnpan-wiht-humann4/9036>\
**Category:** MelonnPan\
**Created:** [August 13, 2026, 7:57pm UTC](https://forum.biobakery.org/t/using-melonnpan-wiht-humann4/9036 "2026-08-13T19:57:32Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Florencia\_Penalba](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/florencia_penalba/32/3704_2.png) [@Florencia\_Penalba](https://forum.biobakery.org/u/Florencia_Penalba)\
**Post date:** [August 13, 2026, 7:57pm UTC](https://forum.biobakery.org/t/using-melonnpan-wiht-humann4/9036/1 "2026-08-13T19:57:32Z")

</div>

I have been using humann v4.0.0.alpha.1 and I have several results. I want to predict the metabolites for this samples. I have tried using melonnpan 0.99.0.predict but it returns an error “No common IDS found between training and test data”. Could this be a case of missmatch of the humann4 output (gene families) with the pretrained data base that package handles?

I don’t have metabolite data to train a new model but I wanted to check if I was doing all the correct steps?