Welcome to

FibeRtypeR

An easy-to-use tool to infer fiber type proportions from bulk human skeletal muscle omics datasets

Read the paper

Data structure

Input data: .csv file

First column should be named 'identifier' containing the gene/protein names. This column name is case sensitive (lower case letters for 'identifier')

Other columns should be the sample names. Do not include any spaces in the sample names

For transcriptomics, gene names or Gene ID's are welcome, which can be specified later

For proteomics, gene names, protein names or UniProt ID's are welcome, which can be specified later

Raw counts (transcriptomics) or LFQ intensities (proteomics) should be provided. fibeRtypeR will take care of the normalization

When ready: proceed to 'Load Data'

Template data

Download template data

Proceed to 'Fiber type deconvolution' tab after data upload'

Raw data

Normalized data

The average prediction error of FibeRtypeR is ±4% for both transcriptome proteome data (based on RMSE values presented in Figure 1 of our paper).

If you use our app, please cite our publication:

Thibaux Van der Stede*, Roger Moreno-Justicia*, Freek Van de Casteele*, Eline Lievens, Alexia Van de Loock, Jonas Vandecauter, Peter Merseburger, Jimmy Van den Eynden, Delphi Van Haver, An Staes, Simon Devos, Morten Hostrup, Pieter Mestdagh,Jo Vandesompele, Ben Stocks, Atul S Deshmukh# & Wim Derave#. 2025 Deconvoluting fiber type proportions from human skeletal muscle transcriptomics and proteomics data using FibeRtypeR. biorXiv, https://doi.org/10.1101/2025.09.17.676778 * Shared first authorship, # Shared corresponding authorship.

Thank you to beta testers Max Ullrich (Ghent University) and Daniel Turner (University of Pavia)

Disclaimer: this tool is designed for RNA sequencing-based transcriptomics or mass spectrometry-based proteomics data

If you have suggestions or having troubles using our tool, please contact us:

wim.derave@ugent.be and atul.deshmukh@sund.ku.dk