Targeted Feature Extraction in MALDI Mass Spectrometry Imaging to Discriminate Proteomic Profiles of Breast and Ovarian Cancer

Abstract

Purpose To develop a mass spectrometry imaging (MSI) based workflow for extracting m/z values related to putative protein biomarkers and using these for reliable tumor classification. Experimental design Given a list of putative breast and ovarian cancer biomarker proteins, a set of related m/z values are extracted from heterogeneous MSI datasets derived from formalin‐fixed paraffin‐embedded tissue material. Based on these features, a linear discriminant analysis classification model is trained to discriminate the two tumor types. Results It is shown that the discriminative power of classification models based on the extracted features is increased compared to the automatic training approach, especially when classifiers are applied to spectral data acquired under different conditions (instrument, preparation, laboratory). Conclusions and clinical relevance Robust classification models not confounded by technical variation between MSI measurements are obtained. This supports the assumption that the classification of the respective tumor types is based on biological rather than technical differences, and that the selected features are related to the proteomic profiles of the tumor types under consideration.

Cancer Tissue Classification Using Supervised Machine Learning Applied to MALDI Mass Spectrometry Imaging

“…For example, deep neural networks have been applied to MALDI MSI data to classify two lung tumor subtypes [7], and non-negative matrix factorization methods to discriminate lung and pancreas tumors [12]. Other examples include classification of six common cancer types [13] and discrimination of breast and ovarian cancer [14] or metastatic breast and pancreatic cancer [15]. Our previous work has adopted supervised ML classification methods to predict the metastatic status of primary EC tissues, achieving 88% accuracy, although sample size was limited [16].…”

Tissue classification by rapid evaporative ionization mass spectrometry (REIMS): comparison between a diathermic knife and CO2 laser sampling on classification performance

“…An ideal situation with real-time tissue classification would be to discriminate different tissue types based on a classification database independently from the handpiece used. Models including tissue-specific peak lists have recently proven to lead to better classification rates [31]. Several classification/prediction models were created based on PCA-LDA with a maximum number of one tissue type and two animals (e.g., cow [liver] vs. chicken [liver]; Fig.…”

“…Differences in the relative intensity of lipid species are included to maximize the variance between two or more groups. Lately, Cordero Hernandez et al reported that the classification rate of tissues can be improved by selecting tissue-specific peaks and generating new models based on both PCA and LDA of acquired molecular profiles [31].…”

Mass Spectrometry Imaging for Reliable and Fast Classification of Non-Small Cell Lung Cancer Subtypes

“…The classification of MSI data has been mainly performed using LDA and SVM [14][15][16][17][18][19][20]. An RF classification algorithm has previously been applied to classify MSI data but not on a large clinically relevant sample cohort [21,22].…”