Skip to main content

R Shiny App Documentation

This section describes the R Shiny applications developed as part of the Plant Genetix toolkit — designed to make genomic and predictive models more accessible to breeders and researchers.

🧰 Current Apps​

1. Genomic Prediction Explorer​

  • Visualize and compare model performance (GBLUP, Random Forest, etc.)
  • Upload new genotype and phenotype datasets
  • Generate GEBV prediction plots and summaries

2. Marker–Trait Explorer​

  • Interactive Manhattan plots and linkage visualization
  • Dynamic filtering by trait, chromosome, or p-value
  • Exportable reports for publication or field decision support

3. Selection Dashboard​

  • Prioritize genotypes or crosses based on multi-trait indices
  • Combine genomic predictions, environmental data, and cost constraints
  • Supports custom weighting schemes and exportable candidate lists

💾 Deployment​

Each Shiny app can be run:

  • Locally using shiny::runApp()
  • On a server via Shiny Server or RStudio Connect
  • In the cloud (e.g., Posit Cloud or Dockerized instance)

🚀 Example Usage​

To launch an app locally, navigate to its directory and run:

shiny::runApp("apps/genomic_prediction_explorer")

Once running, the interface will open in your default browser, allowing you to explore datasets, visualize predictions, and export results interactively.

🧩 Integration​

These Shiny apps are designed to fit seamlessly into the broader Plant Genetix workflow:

  • Predictive Modeling Integration: Load prediction outputs and evaluation metrics directly from model runs.
  • Genomic Selection Datasets: Connect training and validation datasets for interactive exploration.
  • Cross-Language Compatibility: Interface with Python-based models or scripts using the reticulate package.

📘 Additional Resources​