Related Experiment Video
Updated: Nov 21, 2025
08:03
Heuristic Mining of Hierarchical Genotypes and Accessory Genome Loci in Bacterial Populations
Published on: December 7, 2021
2.6K
MegaR: an interactive R package for rapid sample classification and phenotype prediction using metagenome profiles
Eliza Dhungel1, Yassin Mreyoud1, Ho-Jin Gwak2
1Program in Bioinformatics and Computational Biology, Saint Louis University, Saint Louis, MO, USA.
BMC Bioinformatics
|January 19, 2021
Summary
MegaR is a new R Shiny package that simplifies building machine learning models from metagenomic data. It enables accurate microbiome sample classification and phenotype prediction for diverse research applications.
Area of Science:
- Microbiology
- Bioinformatics
- Computational Biology
Background:
- Microbiome communities are crucial for ecosystem functions and animal evolution.
- Metagenomics is powerful for studying microbiome structure and function.
- Understanding microbe-environment relationships is key for human health and ecosystem studies.
Purpose of the Study:
- To develop an accessible tool for building machine learning models from metagenomic data.
- To enable unbiased classification and prediction of microbiome sample properties.
Main Methods:
- Developed MegaR, an R Shiny package and web application.
- Utilized taxonomic profiles from whole metagenome or 16S rRNA sequencing.
- Integrated data processing, machine learning techniques, model validation, and prediction.
Main Results:
- MegaR allows effortless, unbiased machine learning model construction.
- The tool supports classification of samples into multiple categories.
- It offers interactive visualizations and user-friendly model fine-tuning options.
Conclusions:
- MegaR facilitates accurate metagenomic sample classification and phenotype prediction.
- The package aids in identifying and predicting microbe-related human diseases.
- It enables rapid identification of sample properties through unknown sample prediction.
