Li-Chia-ching/Alfalfa_Phenotypic_Clustering_Analysis: v3.0: Alfalfa Phenotype Analysis Tool

We're excited to announce v3.1 of the Alfalfa Clustering Script! This release focuses on improving scientific transparency, visual clarity for publication, and user guidance—without changing the core clustering logic.

Key Improvements

1. Strengthened Scientific Transparency
   While the underlying clustering algorithm remains unchanged, we've made the methodology more explicit and reproducible:

   • Added clear subtitles in the code and output to specify that clustering uses Ward's method on scaled data (essential for handling variables with different units/scales).
   • Automatically generate a detailed interpretation file: 04_Cluster_Interpretation.txt. This file explains the clustering principles, how to interpret the dendrogram, and potential reasons for fragmented or unexpected results (e.g., small sample sizes per group, limited number of variables, environmental noise in phenotypic data, etc.).

2. Publication-Ready Visual Enhancements

   • Unified black branches: In response to feedback, all main tree branches are now standardized to solid black with a thicker line width (1.5 pt). This follows common SCI journal guidelines for clean, distraction-free figures and avoids the visual noise of colored branches.
   • Colorblind-friendly palette (Dark2): Retained the Dark2 qualitative color scheme for the bottom group labels and colored bars. This ensures key cluster/group information remains distinguishable for color-vision-deficient readers while maintaining a professional, low-saturation look.

3. Improved Documentation & Guidance
   The new 04_Cluster_Interpretation.txt file goes beyond basics:

   • Clearly defines what the clusters represent in the context of phenotypic data.
   • Provides actionable next steps if results are suboptimal, such as adding more phenotypic variables, trying alternative distance metrics/algorithms, increasing replication per inbred line, or validating with molecular markers.
     This directly addresses common user questions about "why the tree looks fragmented" and helps bridge analysis to biological interpretation.

These changes make the script more robust for academic workflows, peer-reviewed publications, and reproducible research.

Thanks to everyone who provided feedback—this release directly incorporates your suggestions!

As always, feedback and issues are welcome on the repo. Happy clustering! 🌱