# Transformation of preprocessed metabolomic data for using as MelonnPan input file

**URL:** <https://forum.biobakery.org/t/transformation-of-preprocessed-metabolomic-data-for-using-as-melonnpan-input-file/4392>\
**Category:** MelonnPan\
**Created:** [November 16, 2022, 9:25pm UTC](https://forum.biobakery.org/t/transformation-of-preprocessed-metabolomic-data-for-using-as-melonnpan-input-file/4392 "2022-11-16T21:25:21Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![DEEPCHANDA7](https://avatars.discourse-cdn.com/v4/letter/d/eb9ed0/32.png) [@DEEPCHANDA7](https://forum.biobakery.org/u/DEEPCHANDA7)\
**Post date:** [November 16, 2022, 9:25pm UTC](https://forum.biobakery.org/t/transformation-of-preprocessed-metabolomic-data-for-using-as-melonnpan-input-file/4392/1 "2022-11-16T21:25:21Z")

</div>

Hi community and Dr @himel.mallick !!!  
I have preprocessed raw metabolomic spectra to get an output containing abundance values of each metabolite within a sample in terms of mass spec peak area. However, mass spec experts have suggested performing log transformation on this data prior to any statistical analysis rather than direct TSS normalization which has been performed in the MelonnPan tutorial to get relative abundances.  
I am a bit confused now about how to handle this data file so that it can be used as an input file for the MelonnPan training step. Which transformation should actually I do?

---

<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:** [November 20, 2022, 2:40am UTC](https://forum.biobakery.org/t/transformation-of-preprocessed-metabolomic-data-for-using-as-melonnpan-input-file/4392/2 "2022-11-20T02:40:15Z")

</div>

In the latest version of the [software](https://github.com/biobakery/melonnpan), I have removed the requirement of AST transformation on TSS-normalized metabolites meaning that you should be able to use it for your purpose using alternative normalization and/or transformation (by setting `transform.metab = FALSE`). Please install the latest version of the software and let me know if you encounter any issues.

Thanks a bunch,  
Himel