Peak learning of mass spectrometry imaging data using artificial neural networks

Abstract

Mass spectrometry imaging (MSI) is an emerging technology that holds potential for improving, biomarker discovery, metabolomics research, pharmaceutical applications and clinical diagnosis. Despite many solutions being developed, the large data size and high dimensional nature of MSI, especially 3D datasets, still pose computational and memory complexities that hinder accurate identification of biologically relevant molecular patterns. Moreover, the subjectivity in the selection of parameters for conventional pre-processing approaches can lead to bias. Therefore, we assess if a probabilistic generative model based on a fully connected variational autoencoder can be used for unsupervised analysis and peak learning of MSI data to uncover hidden structures. The resulting msiPL method learns and visualizes the underlying non-linear spectral manifold, revealing biologically relevant clusters of tissue anatomy in a mouse kidney and tumor heterogeneity in human prostatectomy tissue, colorectal carcinoma, and glioblastoma mouse model, with identification of underlying m/z peaks. The method is applied for the analysis of MSI datasets ranging from 3.3 to 78.9 GB, without prior pre-processing and peak picking, and acquired using different mass spectrometers at different centers.

Advances in spatial mass spectrometry enable in-depth neuropharmacodynamics

“…[ 23,87] Trends in Pharmacological Sciences correlation of drugs, metabolites, lipids, peptides, or proteins by MSI with histological and/or pathological features and/or tissue substructures, providing highly relevant complimentary information [17]. Several examples of combining MSI with other imaging modalities, such as confocal Raman microscopy, imaging mass cytometry, magnetic resonance imaging, and positron emission tomography have been reported [17][18][19][20][21].…”

Image to insight: exploring natural products through mass spectrometry imaging

“…This data complexity gives rise to memory and computational challenges, namely “the curse of dimensionality”. 128 Data analysis challenges are further exacerbated when integrating MSI with other imaging modalities. More powerful MSI-specific algorithms are in high demand for more efficient and effective data mining, visualization, image fusion, and interpretation.…”

Uncovering the interference from lipid fragments on the qualification and quantification of serum metabolites in matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometric analysis

“…machine learning) for clinical disease screening. 6,7 To qualify and quantify the metabolites, mass spectrometry (MS) coupled with chromatography (gas chromatography (GC)/MS or high-performance liquid chromatography (HPLC)/MS) analysis is usually employed. 8 The additional separation by chromatography prior to MS analysis could reduce ion suppression effects and ensure the accurate measurement of metabolite concentration.…”

“…Therefore, metabolite analysis is increasingly demanded in large‐scale biological research, especially in the combination of big data‐driven approaches (e.g. machine learning) for clinical disease screening 6,7 …”