A Support Vector Regression Based Model for the Quantitative Prediction of Age and Body Mass Index by using Epigenetic Information from Peripheral Blood

Prediction of human age from epigenetic information can be used to identify human remains for forensic analysis, chronological age, and age related diseases. Aging and obesity contribute to fatal diseases including cancers, circulatory disease, and respiratory disease. Recent studies have proven that CpG dinucleotides are associated with both aging and obesity. In order to identify aging and obesity related CpG dinucleotides computational methods such as feature selection, feature extraction and regression methods can be utilised. In this study, therefore, we examined the methylations level of 27482 CpG dinucleotides in 46 donors from adult peripheral blood to disclose the associations among aging, obesity and epigenetic markers. Body Mass Index (BMI) measure is calculated to decide obesity. To determine aging and obesity related CpG sites, we used unsupervised feature selection methods (USFSMs) as they can be considered as more unbiased approach when compared to the supervised methods. Since USFSMs are independent of any predictive model, support vector regression is applied to evaluate the quality of prediction. Increasing the knowledge of associations among epigenetic mechanisms, aging, and obesity will ultimately allow us to manage and prevent obesity as well as to identify the chronological age of individuals.

[1]  P. Froguel,et al.  The genetics of human obesity , 2005, Nature Reviews Genetics.

[2]  V. Gregory Payne,et al.  The Impact of Obesity and Exercise on Cognitive Aging , 2013, Front. Aging Neurosci..

[3]  Karine Clément,et al.  Genetics of human obesity. , 2006, The Proceedings of the Nutrition Society.

[4]  Deng Cai,et al.  Laplacian Score for Feature Selection , 2005, NIPS.

[5]  Huan Liu,et al.  Embedded Unsupervised Feature Selection , 2015, AAAI.

[6]  J. Mill,et al.  Differential epigenomic and transcriptomic responses in subcutaneous adipose tissue between low and high responders to caloric restriction. , 2010, The American journal of clinical nutrition.

[7]  Jun Ma,et al.  HIF3A DNA Methylation Is Associated with Childhood Obesity and ALT , 2015, PloS one.

[8]  W. Wagner,et al.  Epigenetic-aging-signature to determine age in different tissues , 2011, Aging.

[9]  Marco Cristani,et al.  Infinite Feature Selection , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[10]  B. Christensen,et al.  Aging and Environmental Exposures Alter Tissue-Specific DNA Methylation Dependent upon CpG Island Context , 2009, PLoS genetics.

[11]  Zhoujun Li,et al.  A novel unsupervised feature selection method for bioinformatics data sets through feature clustering , 2008, 2008 IEEE International Conference on Granular Computing.

[12]  Huan Liu,et al.  Spectral feature selection for supervised and unsupervised learning , 2007, ICML '07.

[13]  Xiao Zhang,et al.  Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis , 2010, BMC Bioinformatics.

[14]  Ahmed Bouridane,et al.  A supervised feature selection framework in relation to prediction of antibody feature-function activity relationships in RV144 vaccines , 2016, 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[15]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[16]  Ahmed Bouridane,et al.  Exploration of unsupervised feature selection methods in relation to the prediction of cytokine release effect correlated to antibody features in RV144 vaccines , 2015, 2015 IEEE 15th International Conference on Bioinformatics and Bioengineering (BIBE).

[17]  Jiayi Hou,et al.  Developing a DNA methylation assay for human age prediction in blood and bloodstain. , 2015, Forensic science international. Genetics.

[18]  G. Satten,et al.  Age-associated DNA methylation in pediatric populations. , 2012, Genome research.

[19]  K. Zhao,et al.  Epigenome mapping in normal and disease States. , 2010, Circulation research.

[20]  Wen Liu,et al.  Quantitative prediction of mouse class I MHC peptide binding affinity using support vector machine regression (SVR) models , 2006, BMC Bioinformatics.

[21]  Ahmed Bouridane,et al.  Exploration of unsupervised feature selection methods to predict chronological age of individuals by utilising CpG dinucleotics from whole blood , 2017, 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[22]  R. K. Agrawal,et al.  An Introductory Study on Time Series Modeling and Forecasting , 2013, ArXiv.

[23]  M. H. Juni.  OBESITY: A PUBLIC HEALTH THREATS IN DEVELOPING COUNTRIES , 2015 .

[24]  W. James,et al.  Effect of Plain Versus Sugar‐Sweetened Breakfast on Energy Balance and Metabolic Health: A Randomized Crossover Trial , 2020, Obesity.

[25]  N. Holland,et al.  Relationship between expression and methylation of obesity-related genes in children. , 2015, Mutagenesis.

[26]  M. S. Almén,et al.  Genome-wide analysis reveals DNA methylation markers that vary with both age and obesity. , 2014, Gene.

[27]  C. Ling,et al.  Epigenetics and obesity: the devil is in the details , 2010, BMC medicine.

[28]  B. Korn,et al.  DNA methylation pattern changes upon long-term culture and aging of human mesenchymal stromal cells , 2010, Aging cell.

[29]  J. Mill,et al.  Epigenetics in health and disease: heralding the EWAS era , 2014, The Lancet.