ccrepe

ccrepe evaluates the statistical significance of similarity measures in compositional datasets, with emphasis on microbial community relative-abundance comparisons.

Key Features:

• Similarity measures (`ccrepe`): Calculates similarity measures between relative abundances of microbial communities across one or two body sites using bootstrap and permutation matrices to assess significance.
• Statistical significance: Computes p-values and q-values for similarity measures to control false discovery rate.
• Co-variation and co-exclusion (`nc.score`): Extends the checkerboard score to ordinal data to compute species-level co-variation and co-exclusion patterns.
• Bioconductor integration: Distributed as part of the Bioconductor project for use within the R statistical programming environment.

Scientific Applications:

• Microbiome community comparison: Quantifies and tests similarities between microbial community relative-abundance profiles across body sites or experimental conditions.
• Ecological interaction inference: Identifies species-level co-variation and co-exclusion to infer ecological relationships within microbial communities.
• Impact assessment: Tests hypotheses about the effects of diet, disease, or environmental factors on microbial community structure by evaluating compositional similarity.

Methodology:

Applies bootstrap and permutation testing to generate matrices for computing similarity measures and associated p-values and q-values, and extends the checkerboard score to ordinal data for nc.score-based co-variation and co-exclusion analysis.