MorphoStat

MorphoStat is a modern, interactive R Shiny application for statistical morphological analysis using geometric morphometrics. It provides a comprehensive suite of tools for analyzing landmark-based morphometric data with publication-ready visualizations.

Features

Data Input

• Multiple file formats: Morphologika (.txt), TPS (.tps), NTS (.nts), FCSV (3D Slicer)
• Project save/load: Save and restore complete analysis sessions (.rds)
• Auto-detection: Automatically detects 2D/3D landmark configurations

Analysis Pipeline

• Generalized Procrustes Analysis (GPA): Removes size, position, and rotation effects
• Principal Component Analysis (PCA): Identifies major axes of shape variation
• PERMANOVA: Tests for significant shape differences between groups
• PERMDISP: Tests for homogeneity of group dispersions
• Allometry Analysis: Evaluates shape-size relationships
• Pairwise Comparisons: Multiple group comparisons with distance metrics

Visualizations

• PCA Plots: 2D colored, black & white, and 3D scatter plots
• Shape Deformations: Mean, min, and max PC shape visualization
• Procrustes Superimposition: View all specimens or individual samples
• Interactive 3D WebGL Views: Rotate wireframes in full 3D morphospace using the mouse (Procrustes & PC Deformation)
• Wireframe Diagrams: Multiple anatomical views (dorsal, sagittal, coronal)
• Statistical Plots: Centroid size boxplots, allometry regression, eigenvalue scree plots
• Outlier Detection: Identify potential outliers based on Procrustes distance

Customization

• Custom group colors: Assign specific colors to groups
• Group renaming: Rename groups for publication
• Data spread visualization: Confidence ellipses, convex hulls, density contours
• Diagram transformations: Flip and rotate shape visualizations
• Export options: PNG and SVG formats with customizable dimensions
• Branded interface: Custom logo displayed in the title bar and About section
• Whisker representation: Choose between IQR, Min/Max, Percentile, or Std Deviation for box plots (v1.1)
• Significance bars: Pairwise p-value / star annotations on statistical plots with t-test or Wilcoxon (v1.1)
• Extended plot settings: Custom titles, axis labels, themes, coordinate flip, transparency, and size controls (v1.1)

Installation

Prerequisites

Make sure you have R (>= 4.0.0) installed. You'll also need the following R packages:

install.packages(c(
  "shiny",
  "shinydashboard", 
  "bslib",
  "shinyjs",
  "ggplot2",
  "DT",
  "colourpicker",
  "scatterplot3d",
  "rgl",
  "plotly"
))

# Bioconductor/specialized packages
install.packages("geomorph")
install.packages("RRPP")
install.packages("Morpho")
install.packages("vegan")
install.packages("rstatix")

Running the App

1. Clone this repository:

git clone https://github.com/Dinuka0001/MorphoStat.git
cd MorphoStat

2. Open R or RStudio and run:

shiny::runApp("App.R")

Or run directly:

shiny::runApp("path/to/morphostat")

Usage

Quick Start

1. Select input type: Choose your data format (Morphologika, TPS, NTS, or FCSV)
2. Upload data: Use the file input to upload your landmark data
3. Configure wireframe (optional): Define landmark connections
4. Run analysis: Click "Run Analysis" to perform GPA, PCA, and statistical tests
5. Explore results: Navigate through tabs to view plots and statistics
6. Export: Download results, plots, or save the entire project

Input File Formats

Morphologika Format (.txt)

[individuals]
10
[landmarks]
48
[dimensions]
3
[names]
Specimen_01
...
[labels]
Group_A
...
[rawpoints]
12.345 23.456 5.678
...

TPS Format (.tps)

LM=48
12.345 23.456
13.456 24.567
...
ID=Specimen_01

Wireframe Links

Define landmark connections as comma-separated pairs:

1,2, 2,3, 3,4, 4,5, 5,1

Screenshots

Coming soon

Statistical Methods

Method	Purpose	Package
GPA	Procrustes superimposition	geomorph
PCA	Shape variation analysis	geomorph
PERMANOVA	Group difference testing	vegan, RRPP
PERMDISP	Dispersion homogeneity	vegan
Allometry	Size-shape relationship	RRPP

Citation

If you use MorphoStat in your research, please cite:

Adasooriya, D. (2026). MorphoStat: An Interactive R Shiny Application for
Statistical Morphological Analysis. Version 1.1.0. © 2026 Dinuka Adasooriya.

Also cite the underlying R packages:

• Adams DC, Collyer ML, Kaliontzopoulou A. 2024. Geomorph: Software for geometric morphometric analyses. R package version 4.0.8.
• Oksanen J, et al. 2022. vegan: Community Ecology Package. R package version 2.6-4.

Links

• GitHub: https://github.com/Dinuka0001/MorphoStat
• Online App: https://dinuka-morphostat.share.connect.posit.cloud/

---

Version History

v1.1.0 — April 2026

New Features

Box Plot — Whisker Representation Control

• Added a new Whiskers Representation selector for Centroid Size and Dispersion plots
• Four modes available: Default (1.5× IQR), Min/Max, Percentiles (5th–95th), and Std Deviation (Mean ± SD)
• Uses a statistically correct stat_summary approach — fully compatible with all ggplot2 versions

Box Plot — Significance Bars

• Added pairwise statistical significance annotations to Centroid Size and Dispersion plots
• Choose between t-test or Wilcoxon rank-sum test
• Label style options: star notation (***, **, *, ns) or exact p-values
• Controls for bracket spacing, tip length, and text size
• Significance bars carried through to downloaded plot files

Box Plot — Expanded Aesthetics & Layout

• Fill transparency, box line width, and box width sliders
• Point overlay controls: individual jitter width, alpha, and point size
• Custom axis labels and title for Centroid Size, Dispersion, and Eigenvalues plots
• Theme selector: Classic, Minimal, BW, Light
• Coordinate flip toggle
• Show/hide legend toggle

Eigenvalues Plot — Enhancements

• Customizable bar colour, cumulative line colour
• Bar transparency and width sliders
• Cumulative line width slider
• Theme selector + coordinate flip

2D PCA Plot — Grid Lines Toggle

• Fixed the Show Grid Lines checkbox to correctly show/hide grid lines on 2D scatter plots (both colour and black & white modes)
• Grid state is now preserved in downloaded PNG/SVG exports

Bug Fixes

• SD whisker [object Object]: Rewrote SD whisker logic using stat_summary(fun.data = whisker_stat_fun(...), geom = "boxplot") — the previous aggregate() matrix approach caused serialization errors in Shiny
• Percentile whisker not rendering: geom_boxplot was incorrectly passed fun.data, which it does not support; fixed to use stat_summary consistently across all whisker modes
• PCA grid lines not toggling: morphostat_plot_theme() sets panel.grid.major.x = element_blank() internally; overriding with panel.grid.major was insufficient — fixed by explicitly overriding panel.grid.major.x and panel.grid.major.y individually
• Principal Component Analysis (PCA) — 3D interactive (WebGL) wireframe
• Downloaded PCA plots missing grid: apply_pca_download_theme() used theme_classic() (no grid by default) with only a grid-removal branch — added a grid-addition branch that respects the UI checkbox
• Centroid Size statistics showing raw variable names: aggregate() formula interface was used with reactive values as column names, producing labels like rv$centroid_size in output — replaced with named data frame columns
• Download buttons for statistical plots not working: Handlers download_stat_plot_png, download_stat_plot_svg, and download_all_stat_plots were missing from the server; back-ported from backup

---

v1.0.0 — January 2026 (Initial Release)

• Generalized Procrustes Analysis (GPA) with full landmark superimposition
• Principal Component Analysis (PCA) — 2D colour plot, 2D black & white
• Data spread overlays: confidence ellipses, convex hulls, 2D density contours
• PERMANOVA and PERMDISP group difference testing
• Allometry regression (log centroid size vs PC1)
• Pairwise group comparisons with distance metrics
• Centroid Size, Dispersion, Eigenvalues, and Allometry statistical plots
• Wireframe landmark connection diagrams (dorsal, sagittal, coronal, custom views)
• Shape deformation visualizations (mean, min, max PC)
• Procrustes superimposition overlay plots
• Outlier detection based on Procrustes distance
• Group renaming and custom per-group colour assignment
• Multiple input formats: Morphologika (.txt), TPS (.tps), NTS (.nts), FCSV (3D Slicer)
• Project save/load (.rds)
• Full download suite: PNG, SVG, CSV, ZIP (all statistics), ZIP (all plots)

Author

Dinuka Adasooriya
Yonsei University College of Dentistry

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Acknowledgments

• The R Shiny team for the excellent framework
• The geomorph package authors for their comprehensive morphometrics tools
• The vegan package authors for multivariate analysis tools