Add assay-specific Y-axis labels to plot_predicted_curve()

[Copilot]

plot_predicted_curve() currently hardcodes Y-axis labels as "ELISA units". Different assay types require specific units (optical density, kinetic signal, MFI).

Changes

New parameter: assay (optional, default NULL)

• Accepts "ELISA_OD", "Kinetic_ELISA", "multiplex-bg" (case-insensitive)
• Maps to appropriate Y-axis labels:
  • ELISA_OD → "Optical density (OD)"
  • Kinetic_ELISA → "Kinetic ELISA signal"
  • multiplex-bg → "MFI (background-subtracted)"
• Auto-appends "(log scale)" when log_y = TRUE

Label resolution logic:

1. Use explicit assay parameter if provided
2. Infer from dataset columns (assay, assay_type, assay_units)
3. Fall back to "ELISA units"

Error handling:

• Warns on unsupported assay types, uses default label

Example

# Explicit assay type
plot_predicted_curve(
  model, ids = "patient_123", 
  antigen_iso = "HlyE_IgA",
  assay = "Kinetic_ELISA",
  log_y = TRUE
)
# Y-axis: "Kinetic ELISA signal (log scale)"

# Infer from dataset
dataset$assay <- "ELISA_OD"
plot_predicted_curve(model, ids = "patient_123", antigen_iso = "HlyE_IgA", dataset = dataset)
# Y-axis: "Optical density (OD)"

Backward compatible: existing code unchanged when assay not specified.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• cdn.jsdelivr.net
  • Triggering command: /opt/R/4.5.2/lib/R/bin/exec/R /opt/R/4.5.2/lib/R/bin/exec/R --slave --no-save --no-restore -f /tmp/RtmpO7L4VB/callr-scr-631538b28a65 -B /bin/sh plot_predicted_cbash vignettes/articl/usr/local/bin/quarto (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>add assay-specific labels to plot_predictive_curve.R</issue_title>
<issue_description>pull Y axis labels for the plot depending on the assay units (ELISA OD, Kinetic ELISA, multiplex-bg)</issue_description>

Comments on the Issue (you are @copilot in this section)

@Kwan-Jenny @copilot Please revise `plot_predictive_curve()` in `R/plot_predictive_curve.R` so that the Y-axis label is assay-specific. Add a new optional argument `assay` (or `assay_type`) with default `NULL`. If `assay` is provided, use it to determine the Y-axis label; if it is `NULL`, attempt to infer the assay from the input data (e.g., columns named `assay`, `assay_type`, or `assay_units`), and otherwise fall back to the current default label. Implement a small internal helper that maps assay types (case-insensitive) as follows: `ELISA_OD` → `"Optical density (OD)"`, `Kinetic_ELISA` → `"Kinetic ELISA signal"`, and `multiplex-bg` → `"MFI (background-subtracted)"`; if an unsupported assay is provided, issue a warning and keep the existing label. Update the ggplot call to set the Y-axis label via `labs(y = ...)` without breaking existing behavior. Then update all roxygen2 documentation for `plot_predictive_curve()` to document the new argument and describe the inference and fallback behavior. Add or update `testthat` tests to verify that the returned ggplot object has the correct `y` label for each supported assay and that an unknown assay triggers a warning while preserving the default label. Finally, update any relevant examples, vignettes, or README code that call `plot_predictive_curve()` to include at least one example demonstrating assay-specific labeling, and ensure all tests and examples pass `R CMD check`.

• Fixes add assay-specific labels to plot_predictive_curve.R #176

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.