Open AccessPipeline of Mass-spectrometry Data Processing for Diagnostic Molecular Marker Panel Obtaining Using the Example of Search Markers of Breast Cancer Metastasis
Author(s) -
Alisa Tokareva,
В. В. Чаговец,
Alexey Koikhin,
Natalia Starodubtseva,
Vladimir Frankevich,
Evgeny N. Nikolaev
Publication year - 2021
Publication title -
biomedical chemistry: research and methods
Language(s) - English
Resource type - Journals
ISSN - 2618-7531
DOI - 10.18097/bmcrm00156
Subject(s) - metastasis , pipeline (software) , breast cancer , computational biology , metabolome , computer science , metabolomics , feature selection , selection (genetic algorithm) , proteomics , pathology , bioinformatics , cancer , medicine , biology , artificial intelligence , programming language , biochemistry , gene
A pathology diagnostic using molecular marker is a perspective direction of clinical medicine. Mass-spectrometry (MS) is a one of methods, which are used for obtaining information about molecular profiles. Selection of species, essential for classification “case/control is an important task for data processing. Pipeline of data processing has been proposed using MS data, obtained during analysis of tumor breast tissue samples and health breast tissue samples, with the aim of metastasis marker selection. As a result, selection of lipid markers that belong to classes, related to metastasis and proliferation processes, makes it possible to create high sensitivity diagnostic model, based on logistic regression. The proposed method is applicable for data processing, obtained by MS analysis of other “omics”: metabolome, proteome, glycome.