Building Minimized Epigenetic Clock by iPlex MassARRAY Platform
Genes · 2024
DOI: 10.3390/genes15040425 ↗Kurz gefasst
Fachartikel in der Zeitschrift Genes, 2024. Verfasst von Ekaterina Davydova und 2 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Biology and Biomedicine, Lobachevsky State University, 23 Gagarin Ave., Nizhny Novgorod 603022, Russia. Hauptthemen laut MEDLINE: DNA Methylation, Epigenesis, Genetic, CpG Islands. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 84 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Science and Higher Education of the Russian Federation.
Abstract (Originalsprache)
Epigenetic clocks are valuable tools for estimating both chronological and biological age by assessing DNA methylation levels at specific CpG dinucleotides. While conventional epigenetic clocks rely on genome-wide methylation data, targeted approaches offer a more efficient alternative. In this study, we explored the feasibility of constructing a minimized epigenetic clock utilizing data acquired through the iPlex MassARRAY technology. The study enrolled a cohort of relatively healthy individuals, and their methylation levels of eight specific CpG dinucleotides in genes SLC12A5, LDB2, FIGN, ACSS3, FHL2, and EPHX3 were evaluated using the iPlex MassARRAY system and the Illumina EPIC array. The methylation level of five studied CpG sites demonstrated significant correlations with chronological age and an acceptable convergence of data obtained by the iPlex MassARRAY and Illumina EPIC array. At the same time, the methylation level of three CpG sites showed a weak relationship with age and exhibited a low concordance between the data obtained from the two technologies. The construction of the epigenetic clock involved the utilization of different machine-learning models, including linear models, deep neural networks (DNN), and gradient-boosted decision trees (GBDT). The results obtained from these models were compared with each other and with the outcomes generated by other well-established epigenetic clocks. In our study, the TabNet architecture (deep tabular data learning architecture) exhibited the best performance (best MAE = 5.99). Although our minimized epigenetic clock yielded slightly higher age prediction errors compared to other epigenetic clocks, it still represents a viable alternative to the genome-wide epigenotyping array.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Biology and Biomedicine, Lobachevsky State University, 23 Gagarin Ave., Nizhny Novgorod 603022, Russia.
ORCID-Profile: Ekaterina Davydova, Alexey Perenkov, Maria Vedunova
Förderung
- Ministry of Science and Higher Education of the Russian Federation
Zitieren
APA
Davydova, E., Perenkov, A., & Vedunova, M. (2024). Building Minimized Epigenetic Clock by iPlex MassARRAY Platform. Genes. https://doi.org/10.3390/genes15040425
Vancouver
Davydova E, Perenkov A, Vedunova M. Building Minimized Epigenetic Clock by iPlex MassARRAY Platform. Genes. 2024. doi:10.3390/genes15040425
BibTeX
@article{Davydova2024,
author = {Davydova, Ekaterina and Perenkov, Alexey and Vedunova, Maria},
title = {Building Minimized Epigenetic Clock by iPlex MassARRAY Platform},
journal = {Genes},
year = {2024},
doi = {10.3390/genes15040425}
}
EUVIMED