PERMANOVA family beta diversity by study group

permanova
beta-diversity
family
pre-specified
Bray–Curtis dissimilarity on family relative abundance; study-group effect tested with permutational MANOVA.
Author

IBD Capstone team

Published

May 19, 2026

1 Research question

Does study group explain differences in family-level community composition?

Null: Multivariate centroids do not differ among study groups (PERMANOVA on Bray–Curtis distances).

2 Data

  • Response: Family relative abundance (f__* columns after wide reshape).
  • Predictor: Study_group_new.
  • Input: data/processed/family.csv (long format from make save; pivoted in this post).

3 Methods

Item Choice
Test PERMANOVA (vegan::adonis2)
Distance Bray–Curtis
Formula dist ~ Study_group_new
Permutations 999
Visualization PCoA (cmdscale, k = 2), 95% ellipses

4 Results

Term Df Sum of squares R² F Pr(>F)
Model 2 0.2570 0.1422 1.0773 0.373
Residual 13 1.5505 0.8578 NA NA
Total 15 1.8075 1.0000 NA NA

Study group explained 14.2% of variance in Family composition (R² = 0.142; pseudo-F = 1.08; p = 0.3730, 999 permutations).

5 Figure

Show code 
plot_pcoa_study_group(
  fit$dist,
  fit$metadata,
  permanova_result,
  cfg$label
)
Figure 1: PCoA of family Bray–Curtis dissimilarity by study group.

6 Reproducibility

  • Helpers: stats/R/permanova_helpers.R
  • Pipeline: make wrangle → make save