Related Experiment Video
Updated: Dec 25, 2025
14:56
Sample Preparation to Bioinformatics Analysis of DNA Methylation: Association Strategy for Obesity and Related Trait Studies
Published on: May 6, 2022
5.0K
AgeGuess, a Methylomic Prediction Model for Human Ages.
Xiaoqian Gao1, Shuai Liu1, Haoqiu Song1,2
1BioKnow Health Informatics Laboratory Key Laboratory of Symbolic Computation and Knowledge Engineering, College of Computer Science and Technology, Ministry of Education, Jilin University, Changchun, China.
Frontiers in Bioengineering and Biotechnology
|March 27, 2020
Summary
This study introduces AgeGuess, a new method to predict chronological age using 107 genetic methylation biomarkers. The developed models achieved high accuracy, offering a novel approach to understanding the aging process.
Area of Science:
- Epigenetics and aging research
- Computational biology and bioinformatics
Background:
- Aging is a complex biological process influenced by genetics and molecular changes.
- Methylomic data has shown potential for accurate chronological age prediction.
Purpose of the Study:
- To develop a robust feature selection algorithm for age regression.
- To identify gender-independent methylation biomarkers for age prediction.
Main Methods:
- Proposed a three-step feature selection algorithm named AgeGuess.
- Utilized Support Vector Regressor (SVR) and Ridge regression models.
- Identified 107 gender-independent methylomic features.
Main Results:
- The SVR model achieved a Mean Absolute Deviation (MAD) of 2.0267.
- The Ridge regression model achieved a slightly better MAD of 1.9859.
- Identified only two shared methylation biomarkers between gender-specific models, located in CALB1 and KLF14 genes.
Conclusions:
- The AgeGuess algorithm effectively identifies age-predictive methylomic biomarkers.
- Gender-independent models provide a reliable method for age prediction.
- Further improvements are possible with gender-specific models, highlighting the nuanced role of specific biomarkers in aging.
