Application of Machine Learning in Spatial Proteomics

Abstract

Spatial proteomics is an interdisciplinary field that investigates the localization and dynamics of proteins, and it has gained extensive attention in recent years, especially the subcellular proteomics. Numerous evidence indicate that the subcellular localization of proteins is associated with various cellular processes and disease progression. Mass spectrometry (MS)-based and imaging-based experimental approaches have been developed to acquire large-scale spatial proteomic data. To allow the reliable analysis of increasingly complex spatial proteomics data, machine learning (ML) methods have been widely used in both MS-based and imaging-based spatial proteomic data analysis pipelines. Here, we comprehensively survey the applications of ML in spatial proteomics from following aspects: (1) data resources for spatial proteome are comprehensively introduced; (2) the roles of different ML algorithms in data analysis pipelines are elaborated; (3) successful applications of spatial proteomics and several analytical tools integrating ML methods are presented; (4) challenges existing in modern ML-based spatial proteomics studies are discussed. This review provides guidelines for researchers seeking to apply ML methods to analyze spatial proteomic data and can facilitate insightful understanding of cell biology as well as the future research in medical and drug discovery communities.

Navigating the Multiverse: A Hitchhiker’s Guide to Selecting Harmonisation Methods for Multimodal Biomedical Data

“…Spectral data, typically acquired through mass spectrometry experiments to study protein molecules and metabolites, provides detailed insights into the structural composition, constitution and organisation of the molecules under investigation (Mansuri et al (2023); Mou et al (2022)). ML analysis of spectral data involves features representing three-dimensional conformations and spatial relationships of molecules, enabling classification based on functional groups and elements (Mou et al (2022); Sachdev and Gupta (2019)).…”

“…Spectral data, typically acquired through mass spectrometry experiments to study protein molecules and metabolites, provides detailed insights into the structural composition, constitution and organisation of the molecules under investigation (Mansuri et al (2023); Mou et al (2022)). ML analysis of spectral data involves features representing three-dimensional conformations and spatial relationships of molecules, enabling classification based on functional groups and elements (Mou et al (2022); Sachdev and Gupta (2019)). Proteomics, examining proteins through expression, functional relationships, and structural information, includes investigations into protein folding and structural orientations using methods such as NMR and X-ray crystallography (Malet-Martino and Holzgrabe (2011)).…”

Evaluating Machine Learning Methods of Analyzing Multiclass Metabolomics

“…For example, more methods (such as glmnet ridge regression) can be used to impute these missing values to select the most appropriate imputation method . Other deep learning algorithms (such as deep neural networks) are competitive methods for supervised classification of multiclass metabolomics. , Second, these machine learning methods for processing multiclass metabolomic data can be assessed using more different evaluation criteria. A comprehensive assessment of these methods is beneficial for selecting the appropriate analytical process.…”

Progress in research on tumor microenvironment-based spatial omics technologies

“…Commonly used spatial proteomics technology is based on DSP, which integrates the quantitative information of the proteome with in situ tissue information, enabling in situ coanalysis of up to hundreds of proteins on a single paraffin tissue section. Unlike multiplexed immunohistochemistry and multiplexed immunofluorescence, spatial proteomics technology captures the target protein in situ using an antibody coupled to the nucleic acid probe, followed by the release of the nucleic acid probe through special photodissociation [21]. It uses the nCounter digital label technology for counting and quantification, thereby directly reflecting the abundance of the target protein.…”