The Omics Molecule Extractor: A Web Application for the Selection of Potential Biomarker Panels

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Selecting molecular panels for patient health classification is crucial in omics data analysis.
• Current software often requires programming skills, lacks user-friendliness, or cannot handle large datasets.

Purpose of the Study:

• To introduce the Omics Molecule Extractor (OMEx), an open-source web application for user-friendly molecule and panel selection from large omics datasets.
• To provide an accessible tool for identifying biomarkers for sample classification.

Main Methods:

• Developed OMEx, a web application with an interactive user interface for data exploration and analysis.
• Implemented a machine learning-based feature selection strategy for identifying group-separating molecules.
• Utilized interactive visualizations for exploring input data and analysis results.

Main Results:

• The algorithm demonstrated high precision in discovering group-separating molecules on synthetic datasets with known ground truth.
• Testing on five real-world omics datasets showed high reproducibility and overlap with existing methods.
• OMEx also identified alternative molecules of potential interest.

Conclusions:

• OMEx offers a user-friendly, machine learning-driven solution for molecule selection in omics data analysis.
• The tool enhances the ability to classify sample groups and identify relevant biomarkers.
• OMEx is freely available, promoting accessibility for researchers in bioinformatics and related fields.