dar

dar implements consensus-based differential abundance analysis for microbiome data by enabling dplyr-like pipeable sequences of multiple differential abundance methods with subsequent consensus strategies.

Key Features:

• Pipeable workflow: Supports dplyr-like pipeable sequences of differential abundance methods.
• Consensus strategies: Applies different consensus strategies to aggregate results from multiple differential abundance methods.
• Multiple method integration: Combines parametric and non-parametric statistical methods to address sparsity, high variability, and compositional nature of microbiome data.

Scientific Applications:

• Microbiome differential abundance testing: Identifies differentially abundant taxa in microbiome data while balancing specificity and sensitivity to reduce type I and type II errors.
• Robust statistical inference: Increases reliability of differential abundance results by leveraging consensus across multiple methods rather than relying on a single approach.

Methodology:

The tool sequences multiple differential abundance methods in a pipeable workflow and applies consensus strategies to their outputs; specific algorithmic details of individual methods or consensus functions are not explicitly stated in the sources.