MetaAnalyst is a free, user-friendly software package for metagenomic biomarker detection and phenotype classification. MetaAnalyst provides the following functionalities as an efficient analysis tool for metagenomic biomarker detection and phenotype classification:
- Accepts seven different types of input files.
- Supports multilevel labeling that enables researchers to analyze the data from different perspectives (i.e., under various conditions) smoothly.
- Provides a variety of pre-processing procedures before downstream statistical analysis.
- Includes about 28 biomarker detection algorithms and 4 classifiers
- Provides three criteria for evaluating the performance of biomarker detection algorithms:
- Supervised classification performance: MetaAnalyst computes the overall classification accuracy (ACC), balanced accuracy (BACC), sensitivity (SEN), specificity (SPC), miss classification rate (MCR), receiver operation curve (ROC), and area under the curve (AUC).
- Unsupervised clustering performance
- Consensus performance
- Generate the output in tab-delimited files and publication quality plots with various formatting capabilities.
The MetaAnalyst software is implemented using Matlab R2021a and it is available freely as a stand-alone package for Windows operating system at (https://gjuedujo-my.sharepoint.com/:u:/g/personal/mustafa_shawaqfeh_gju_edu_jo/EckO_P5LVyhOh9w-RXisqQMBmIhKLSutGEjjXZOPP9_2fg?e=HGkR0m)
MetaAnalyst is a standalone desktop application, written in MATLAB and does not require any additional packages to install. MetaAnalyst runs on Microsoft Windows.