Dynamic Data Driven Application Systems for Identification of Biomarkers in DNA Methylation

摘要: The term ‘epigenetic’ refers to all heritable alterations that occur in a given gene function without having any change on the DeoxyriboNucleic Acid (DNA) sequence. Epigenetic modifications play crucial role development and differentiation of various diseases including cancer. specific epigenetic alteration has garnered great deal attention is DNA methylation, i.e., addition methyl-group cytosine. Recent studies have shown different tumor types distinct methylation profiles. Identifying idiosyncratic profiles subtypes can provide invaluable insights for accurate diagnosis, early detection, tailoring related treatment In this study, our goal identify informative genes (biomarkers) whose level correlates with cancer type or subtype. To achieve goal, we propose novel high dimensional learning framework inspired by dynamic data driven application systems paradigm biomarkers, determine outlier(s) improve quality resultant disease detection. proposed starts principal component analysis (PCA) followed hierarchical clustering (HCL) observations determination based HCL predictions. capabilities performance are demonstrated using dataset stored Gene Expression Omnibus (GEO) DataSets lung preliminary results demonstrate outperforms conventional algorithms embedded dimension reduction methods, its efficiency outliers, removal their contaminating effects at expense reasonable computational cost.