Cover Image

Enhancing mass spectrometry imaging accessibility using convolutional autoencoders for deriving hypoxia-associated peptides from tumors

Mass spectrometry imaging (MSI) allows to study cancer’s intratumoral heterogeneity through spatially-resolved peptides, metabolites and lipids. Yet, in biomedical research MSI is rarely used for biomarker discovery. Besides its high dimensionality and multicollinearity, mass spectrometry (MS) techn...

Full description

Saved in:
Bibliographic Details
Main Authors: Bitto, Verena (Author) , Hönscheid, Pia (Author) , Besso, María José (Author) , Sperling, Christian (Author) , Kurth, Ina (Author) , Baumann, Michael (Author) , Brors, Benedikt (Author)
Format: Article (Journal)
Language:English
Published: 27 May 2024
In: npj Systems biology and applications
Year: 2024, Volume: 10, Issue: 1, Pages: 1-13
ISSN:2056-7189
DOI:10.1038/s41540-024-00385-x
Online Access:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1038/s41540-024-00385-x
Verlag, lizenzpflichtig, Volltext: https://www.nature.com/articles/s41540-024-00385-x
Get full text
Author Notes:Verena Bitto, Pia Hönscheid, María José Besso, Christian Sperling, Ina Kurth, Michael Baumann, Benedikt Brors
Description
Summary:Mass spectrometry imaging (MSI) allows to study cancer’s intratumoral heterogeneity through spatially-resolved peptides, metabolites and lipids. Yet, in biomedical research MSI is rarely used for biomarker discovery. Besides its high dimensionality and multicollinearity, mass spectrometry (MS) technologies typically output mass-to-charge ratio values but not the biochemical compounds of interest. Our framework makes particularly f num low-abundant signals in MSI more accessible. We utilized convolutional autoencoders to aggregate features associated with tumor hypoxia, a parameter with significant spatial heterogeneity, in cancer xenograft models. We highlight that MSI captures these low-abundant signals and that autoencoders can preserve them in their latent space. The relevance of individual hyperparameters is demonstrated through ablation experiments, and the contribution from original features to latent features is unraveled. Complementing MSI with tandem MS from the same tumor model, multiple hypoxia-associated peptide candidates were derived. Compared to random forests alone, our autoencoder approach yielded more biologically relevant insights for biomarker discovery.
Item Description:Gesehen am 11.06.2025
Physical Description:Online Resource
ISSN:2056-7189
DOI:10.1038/s41540-024-00385-x

Similar Items