MECAF
MECAF performs differential abundance testing on high-dimensional microbiome compositional data to detect condition-associated changes without assuming equal covariance across groups.
Key Features:
- Centered Log-Ratio Transformation: Employs centered log-ratio (CLR) transformation to analyze microbiome compositions and compare two high-dimensional abundance data matrices under different conditions.
- Closed-Form P-value Computation: Computes test p-values directly using a closed-form solution derived from the asymptotic null distribution.
- Maximum-Type Equal-Covariance-Assumption-Free Test: Implements a maximum-type test that operates without assuming equal covariance across groups.
- Asymptotic Power Analysis Against Sparse Alternatives: Investigates asymptotic statistical power specifically against sparse alternatives common in microbiome studies.
- Controlled Type I Error via Simulations: Demonstrates well-controlled type I error rates and improved statistical power relative to competing methods through simulation studies.
- Compatibility with High-Throughput Metagenomic Sequencing: Applicable to abundance data derived from high-throughput metagenomic sequencing technologies.
Scientific Applications:
- Differential Abundance Discovery: Detects shifts in microbiome abundances between disease and treatment conditions.
- Clinical Pathogen and Infectious Disease Studies: Supports analyses for clinical pathogen diagnosis and studies of infectious disease using metagenomic abundance data.
- Real-World Microbiome Data Analysis: Applied to real-world microbiome datasets to assess condition-associated microbial dynamics.
Methodology:
Computational steps explicitly include centered log-ratio transformation of compositional abundance matrices, a maximum-type equal-covariance-assumption-free test statistic, closed-form p-value derivation from the asymptotic null distribution, asymptotic power analysis against sparse alternatives, and evaluation via simulation studies.
Topics
Details
- License:
- GPL-3.0
- Cost:
- Free of charge
- Tool Type:
- command-line tool
- Operating Systems:
- Mac, Linux, Windows
- Programming Languages:
- R
- Added:
- 2/8/2023
- Last Updated:
- 11/24/2024
Operations
Publications
Li Z, Yu X, Guo H, Lee T, Hu J. A maximum-type microbial differential abundance test with application to high-dimensional microbiome data analyses. Frontiers in Cellular and Infection Microbiology. 2022;12. doi:10.3389/fcimb.2022.988717. PMID:36389165. PMCID:PMC9650337.