Home Medizin Modelle für maschinelles Lernen zur Vorhersage des Alters anhand von Veränderungen des Gehirntranskriptoms

Modelle für maschinelles Lernen zur Vorhersage des Alters anhand von Veränderungen des Gehirntranskriptoms

von NFI Redaktion

On the cover of a new research paper published in Aging (listed by MEDLINE/PubMed as „Aging (Albany NY)“ and „Aging-US“ by Web of Science), Volume 16, Issue 5, titled „Genome-wide Transcriptome Profiling and Development of Age Prediction Models in the Human Brain.“

Prior studies have examined age-related transcriptome changes in various regions of the healthy human brain. However, a study on developing age prediction models based on expression levels of specific transcript groups has been lacking. Additionally, studies investigating sexually dimorphic gene activities in the aging brain have reported conflicting results, suggesting the need for additional research. Previous studies comparing different brain regions in the human brain have shown that the prefrontal cortex (PFC) region exhibits a high number of significant transcriptome changes during healthy aging.

In this new study, researchers Joseph A. Zarrella and Amy Tsurumi from Harvard TH Chan School of Public Health, Massachusetts General Hospital, Harvard Medical School, and the Shriner’s Hospitals for Children-Boston aimed to profile PFC transcriptome changes during healthy human aging overall and compare potential differences between female and male samples, as well as develop chronological age prediction models using various techniques.

„We harmonized neuropathologically normal PFC transcriptome datasets from the Gene Expression Omnibus (GEO) repository across ages 21 to 105 and found a large number of differentially regulated transcripts in older individuals compared to younger samples. We compared female and male-specific expression changes.“

The team evaluated age-associated genes using ontology, pathway, and network analyses. They also utilized established (Least Absolute Shrinkage and Selection Operator (Lasso) and Elastic Net (EN)) and newer (eXtreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM)) machine learning algorithms to develop accurate chronological age prediction models and validated them. Further studies to validate these models in other large populations and molecular studies to elucidate the potential mechanisms through which the identified transcripts may relate to aging phenotypes would be beneficial.

„Our results support the notion that specific gene expression changes in the PFC strongly correlate with age, some transcripts exhibit female and male-specific differences, and machine learning algorithms are useful tools for developing age prediction models based on transcriptomic information.“

Source:

Journal Reference:

Zarrella, JA, & Tsurumi, A. (2024). Genome-wide Transcriptome Profiling and Development of Age Prediction Models in the Human Brain. Aging. doi.org/10.18632/aging.205609.

Related Posts

Adblock Detected

Please support us by disabling your AdBlocker extension from your browsers for our website.