Analysis Log

Reproducible Framework for Multi-Omics Analysis in IBD

This site documents every statistical analysis in the project: the method and software settings, tabulated results, interpretation, and the plots that go with each analysis.

Note

All results on this site are computed from synthetic data. They show how the analyses work and are not real study findings.

1 Team

Team: Tiffany Chu, Victoria Farkas, Ian Gault, Derrick Jaskiel
Mentor: Payman Nickchi

2 Where to find analyses

Tab	YAML category	Contents
Exploratory Analysis	`eda`	Heatmaps and descriptive pipeline figures (no hypothesis test)
Pre-specified Tests	`pre-specified`	PERMANOVA, Kruskal–Wallis, Spearman, PCA/clustering, evidence-ranking matrix

The home page lists all posts. Use the navbar tabs to filter by type.

3 Initial exploration notebooks

Early-stage notebooks in notebooks/ are preserved for reference.

Notebook	Contents
`notebooks/stool_eda.Rmd`	Stool mycobiome EDA, alpha diversity, Dunn post-hoc
`notebooks/permanova.Rmd`	PERMANOVA and PCoA development
`notebooks/pca_tsne.Rmd`	PCA and t-SNE exploration
`notebooks/mycobiome_interactive.Rmd`	Interactive abundance visualizations

Analysis	Source
PERMANOVA + PCoA (phylum, family, genus, species)	`stats/posts/2026-05-19-permanova-*`, `stats/R/permanova_helpers.R`, `notebooks/permanova.Rmd`
Symptoms × fungal composition (Spearman + PERMANOVA)	`stats/posts/2026-05-22-symptoms-fungal-composition/`, `stats/R/symptom_association_helpers.R`
Nutrients × diversity by disease group (Spearman)	`stats/posts/2026-05-29-nutrients-diversity-disease-group/`, `stats/R/nutrient_association_helpers.R`
Diet + mycobiome PCA / clustering	`stats/posts/2026-05-29-diet-mycobiome-pca-clustering/`, `stats/R/pca_clustering_helpers.R`, `notebooks/pca_tsne.Rmd`
Evidence-ranking matrix (disease group)	`stats/posts/2026-05-29-evidence-ranking-disease-group/`, `stats/R/evidence_ranking_helpers.R`
Kruskal–Wallis (alpha diversity)	`src/mycobiome/03b_diversity_stats.R`, `notebooks/stool_eda.Rmd`
PCA / t-SNE	`notebooks/pca_tsne.Rmd`

4 How to add a new analysis

Copy stats/_template.qmd into a new folder, e.g. stats/posts/2026-05-19-my-test/index.qmd.
Fill in the YAML (title, date, categories, description) — tag with eda or pre-specified as appropriate.
Fill in each section: research question → data → methods → results → plot.
Render from the repo root:
```
quarto render stats
```
Or use Render in RStudio / Positron.

Participant characteristics: missingness audit

eda

characteristics

Descriptive audit of missing values in the cleaned participant characteristics file, including variable-level rates, participant-level burden, and core clinical coverage.

Jun 16, 2026