patternize: An R package for quantifying colour pattern variation.

Area of Science:

• Ecology and Evolutionary Biology
• Bioinformatics
• Computational Biology

Background:

• Quantifying color patterns in organisms is crucial for understanding variation.
• Existing methods often lack standardized approaches for image alignment and segmentation.
• Image analysis requires establishing homology and segmenting colors for accurate quantification.

Purpose of the Study:

• To introduce patternize, an R package for image alignment and segmentation.
• To provide a tool for quantifying color pattern variation in diverse organisms.
• To facilitate comparative analyses of color patterns in biological research.

Main Methods:

• Patternize utilizes manual landmarks or automated registration for image alignment and homology.
• Color pattern identification employs RGB thresholding, k-means clustering, or watershed transformation.
• The package integrates with principal component analysis (PCA) for data visualization and analysis.

Main Results:

• Patternize successfully quantifies color patterns in butterflies, guppies, spiders, and salamanders.
• The package allows for comparison, visualization (heatmaps), and statistical analysis of image data.
• Demonstrated utility in analyzing fine-scale color pattern phenotypes.

Conclusions:

• Patternize offers a versatile solution for quantifying color pattern variation.
• The R package supports applications in population genetics, evolutionary studies, and phenotype analysis.
• Enables detailed investigation into the genetic and environmental basis of color patterns.