Outlier Detection (plotOutliers)
This analysis identifies potential outliers in the Procrustes-aligned coordinates using Procrustes distance from the mean shape.Outlier Summary:
Settings
Procrustes Plot Settings
PC Deformation
PC Axis Direction
Diagram Transformations:
Display Options:
Colors:
Download
Outlier Detection Settings
Outliers are specimens whose Procrustes distance from the mean shape exceeds a threshold based on standard deviations.Colors:
Download
Data Summary
PC Variance Explained
Centroid Size Analysis
Allometry Regression
PERMANOVA Results (Shape-based)
PERMANOVA on 95% PCs (Euclidean Distance)
PERMDISP Test (Homogeneity of Dispersions)
Pairwise Comparisons
Group Centroids
Plot Settings
Centroid Size Box Plot
Group Colors
Font Sizes
Allometry Plot
Colors
Font Sizes
Shape Vectors:
Font Sizes:
Dispersion Plot:
Group Colors:
Box Styling:
Font Sizes:
Eigenvalues Plot:
Font Sizes:
Download Current Plot
Download All Plots
Dorsal View (Top View - XY Plane)
Sagittal View (Side View - XZ Plane)
Coronal View (Frontal View - YZ Plane)
2D Shape Visualization
TPS Transformation Grid
Wrapped Outline Drawing
Landmark Displacement Graph
This graph shows the displacement magnitude of each landmark from the mean shape to the minimum and maximum PC scores.Settings
PC Deformation Settings
PC Axis Direction
Diagram Transformations
Display Options
Shape Colors:
Download
PC Scores Data
Principal component scores for each specimen, showing position in morphospace.
Download PC Scores (.csv)
Average (Mean) Shape Coordinates
Landmark coordinates of the consensus/mean shape from Procrustes superimposition.
Download Average Shape (.csv)
Principal Component Coefficients (Loadings)
Eigenvectors showing how each landmark coordinate contributes to each PC.
Download PC Coefficients (.csv)
Centroid Size Data
Centroid size (overall size measure) for each specimen.
Download Centroid Size (.csv)
Procrustes-Aligned Coordinates
Full Procrustes-superimposed landmark coordinates for all specimens.
Download Procrustes Coordinates (.csv)
Variance Explained by PCs
Proportion of total shape variance explained by each principal component.
Download Variance Table (.csv)
Getting Started
- Select your input data type (Morphologika, TPS, or NTS)
- Upload your data file using the file input
- Configure wireframe links if needed
- Customize colors and analysis settings
- Click 'Run Analysis' to process your data
Input Formats
Morphologika (.txt)
[individuals] 10 [landmarks] 48 [dimensions] 3 [names] Specimen_01 Specimen_02 ... [labels] Group_A Group_A ... [rawpoints] 12.345 23.456 5.678 13.456 24.567 6.789 ...
Contains metadata headers followed by 3D/2D coordinates
TPS (.tps)
LM=48 12.345 23.456 13.456 24.567 14.567 25.678 ... ID=Specimen_01 IMAGE=image01.jpg LM=48 15.678 26.789 16.789 27.890 ...
Each specimen starts with LM=n (landmark count), followed by coordinates
NTS (.nts)
1 10L 48 3 0 Specimen_01 12.345 23.456 5.678 13.456 24.567 6.789 ... Specimen_02 15.678 26.789 7.890 ...
Header line: 1=rectangular matrix, nL=specimens, landmarks, dimensions, 0=no missing
FCSV (.fcsv)
# Markups fiducial file version = 4.11 # columns = id,x,y,z,ow,ox,oy,oz,vis,sel,lock,label,desc,associatedNodeID vtkMRMLMarkupsFiducialNode_0,-12.5,34.2,56.7,0,0,0,1,1,1,0,LM1,, vtkMRMLMarkupsFiducialNode_1,-10.3,32.1,54.5,0,0,0,1,1,1,0,LM2,, vtkMRMLMarkupsFiducialNode_2,-8.7,30.5,52.3,0,0,0,1,1,1,0,LM3,, ...
3D Slicer fiducial markup format. Each row is a landmark with id, x, y, z coordinates and metadata. Multiple FCSV files (one per specimen) should be provided.
Wireframe Links
Define landmark connections as comma-separated pairs:
1,2, 2,3, 3,4, ...
Each pair represents two landmarks to be connected in the wireframe.
Features
- PCA Visualization: 2D colored/B&W plots and 3D interactive plots
- PC Deformations: Mean/min/max shape views with anatomical projections
- Statistical Analysis: PERMANOVA (shape and PC-based), PERMDISP
- Statistical Plots: Centroid size boxplots, allometry, eigenvalues
- Procrustes Visualization: View all specimens or individual samples
- Customization: Group renaming, custom colors, multiple export formats
Statistical Methods
This application implements robust geometric morphometric workflows:
- GPA Alignment: Generalized Procrustes Analysis removes size, position, and rotation effects
- PCA: Principal Component Analysis identifies major axes of shape variation
- PERMANOVA: Permutational MANOVA tests for significant shape differences
- PERMDISP: Tests for homogeneity of group dispersions
- Allometry: Procrustes ANOVA evaluates shape-size relationships
MorphoStat
v1.0
Statistical Morphological Analysis
An interactive tool for statistical comparison and visualization of morphological measurements.
Developed by Dinuka Adasooriya
Yonsei University College of Dentistry
R Packages Used
shiny, bslib, shinyjs
geomorph, Morpho, RRPP
ggplot2, scatterplot3d, rgl
vegan, rstatix, DT
Citation
If you use MorphoStat in your research, please cite the relevant 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.
2025 Dinuka Adasooriya. All rights reserved.