Applications of machine learning in metabolomics: Disease modeling and classification

Abstract

Metabolomics research has recently gained popularity because it enables the study of biological traits at the biochemical level and, as a result, can directly reveal what occurs in a cell or a tissue based on health or disease status, complementing other omics such as genomics and transcriptomics. Like other high-throughput biological experiments, metabolomics produces vast volumes of complex data. The application of machine learning (ML) to analyze data, recognize patterns, and build models is expanding across multiple fields. In the same way, ML methods are utilized for the classification, regression, or clustering of highly complex metabolomic data. This review discusses how disease modeling and diagnosis can be enhanced via deep and comprehensive metabolomic profiling using ML. We discuss the general layout of a metabolic workflow and the fundamental ML techniques used to analyze metabolomic data, including support vector machines (SVM), decision trees, random forests (RF), neural networks (NN), and deep learning (DL). Finally, we present the advantages and disadvantages of various ML methods and provide suggestions for different metabolic data analysis scenarios.

Metabolic insights into tumor pathogenesis: Unveiling pan‐cancer metabolism and the potential of untargeted metabolomics

“…Supervised learning methods mainly include linear regression, logistic regression, Linear Discriminant Analysis (LDA), k‐Nearest Neighbers (k‐NN), Support Vector Machines (SVM), decision tree and Random Forest (RF), partial least‐squares discrimination analysis (PLS‐DA), orthogonal partial least‐squares discrimination analysis (OPLS‐DA), neural networks (NN), and other methods. Examples of unsupervised algorithms include principal component analysis (PCA), kernel PCA, t‐distributed stochastic neighbor embedding (t‐SNE), uniform manifold approximation and projection (UMAP), k‐means Clustering, apriori, hierarchical clustering, and so on 187 . In the study of early detection of lung adenocarcinoma and classification of lung nodules, Wang et al 191 utilized the t‐SNE method to perform dimensionality reduction analysis on metabolic data.…”

“…( 2 8) Model optimization: Optimize the model based on the actual effect and repeat the above steps (Figure 3). 187 As is well known, metabolic data need to be standardized before statistical analysis to eliminate differences between different batches and instruments, including peak extraction, peak detection, peak alignment, missing value filling, metabolite annotation, and other steps. DL methods have been proposed for critical data preprocessing steps.…”

“…(8) Model optimization: Optimize the model based on the actual effect and repeat the above steps (Figure 3). 187 …”

Plasma Instead of Serum Avoids Critical Confounding of Clinical Metabolomics Studies by Platelets

“…1 In contrast to other biomolecules, metabolites may be formed and degraded by a variety of different and independent mechanisms, rendering data interpretation more difficult and calling for supportive machine learning algorithms. 2 However, metabolomics data may improve the diagnosis of diseases, help to better understand disease mechanisms, and represent an important tool to practice precision medicine supporting individualized drug treatments and monitoring therapeutic outcomes. 3 Clinical metabolomics is typically performed using serum or plasma as sample matrices.…”

“…Metabolomics represents a contemporary postgenomic analysis method for small molecules, comprising building blocks for biosynthesis, fuel for energy production along with the corresponding waste products as well as catalytically active metabolites and signaling molecules . In contrast to other biomolecules, metabolites may be formed and degraded by a variety of different and independent mechanisms, rendering data interpretation more difficult and calling for supportive machine learning algorithms . However, metabolomics data may improve the diagnosis of diseases, help to better understand disease mechanisms, and represent an important tool to practice precision medicine supporting individualized drug treatments and monitoring therapeutic outcomes …”