---
title: "From IMED evidence to a defensible manuscript analysis"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{From IMED evidence to a defensible manuscript analysis}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, eval=FALSE}
library(imedR)
db <- imed_data()

# 1. Audit the frozen release.
imed_validate(db)

# 2. Create one defensible comparison set.
x <- imed_analysis_set(
  db,
  comparability_class = "CMP-W-VOL",
  metric_type = "Abundance"
)

# 3. Describe and visualize without mixing incompatible denominators.
imed_describe(x)
p <- imed_plot(x, type = "histogram")
imed_export_figure(p, "figure1.tiff", dpi = 600)

# 4. Check whether formal evidence synthesis is justified.
imed_meta_eligibility(x)

# 5. Retrieve the exact primary-study citation/provenance trail.
imed_cite(db, unique(x$Study_ID))
imed_provenance(db, study_id = unique(x$Study_ID))
```

`imedR` treats refusal as a scientific output. If a subset mixes comparability
classes, unresolved wet/dry sediment basis, incompatible units, or non-independent
effects, strict synthesis is blocked until the researcher resolves the design.