Differential Expression, Functional and Machine Learning Analysis of High-Throughput -Omics Data Using Open-Source...

Area of Science:

• Genomics and Bioinformatics
• Oral Biology
• Biomedical Data Analysis

Background:

• High-throughput omics analyses (RNA, microRNA, DNA methylation) offer comprehensive genome-level insights into complex diseases.
• Traditional statistical methods struggle with the high-dimensional data generated by omics studies.
• Omics technology has been successfully applied to identify key molecules and pathways in periodontal disease.

Purpose of the Study:

• To present a robust bioinformatics workflow for the initial analysis of omics data.
• To outline a differential expression analysis pipeline suitable for various omics experiments.
• To introduce machine learning approaches for advanced pattern detection and biological inference.

Main Methods:

• Utilized open-source bioinformatics tools for data analysis.
• Developed a differential expression analysis pipeline accounting for fixed or random effects.
• Applied supervised classification and unsupervised clustering for functional analysis and class discovery.

Main Results:

• The workflow enables unbiased, comprehensive genome-level analysis of complex diseases.
• Identified key molecules and pathways in periodontal disease using omics technology.
• Machine learning facilitated pattern detection beyond simple differential feature identification.

Conclusions:

• The presented bioinformatics workflow provides a powerful approach for analyzing high-dimensional omics data.
• This methodology enhances the biological inference from omics studies, particularly in complex diseases.
• The integration of machine learning offers advanced capabilities for class validation and discovery in omics research.