Human Phylogenetic Tree and Genetic Similarity Analysis using Short Tandem Repeats (STR)

This repository combines two key analyses on genetic data using Short Tandem Repeats (STR):

1. Ward Linkage Hierarchical Clustering: Constructs a phylogenetic tree to visualize genetic relationships between individuals.
2. Genetic Similarity & Distance Matrices: Calculates and visualizes multiple similarity and distance metrics between individuals based on their genetic profiles.

Description

• Hierarchical Clustering (Ward Linkage): Uses Ward's method to perform agglomerative clustering, minimizing variance within clusters. The result is displayed as a dendrogram, where the distance between branches reflects genetic dissimilarity.
• Similarity/Distance Metrics: Computes and visualizes the following matrices:
  • Cosine Similarity Matrix
  • Euclidean Distance Matrix
  • Pearson Correlation Matrix
  • Spearman Correlation Matrix

Both analyses are designed to infer genetic relationships, visualize similarity, and highlight genetic variance between samples.

Features

• Hierarchical Clustering: Constructs a dendrogram annotated with genetic distances between individuals.
• Similarity & Distance Matrices: Generates and visualizes key metrics such as cosine similarity, Euclidean distance, and Pearson/Spearman correlations.

Requirements

• pandas
• numpy
• scikit-learn
• scipy
• matplotlib
• seaborn

Usage

1. CSV File: Ensure your genetic data is in a CSV file format.
   • Example : synthatic_sTR.csv. Note that this contains synthatic sTR data for demonstration purposes.
2. Change File Paths: Update the file paths in the .py scripts to match the location of your .csv files.

file_path = 'your_file.csv'  # Replace with your file path

Usage

python clustering_script.py

Running the Clustering Script

python clustering_script.py

This will generate a dendrogram saved as gene_hierarchy_with_distances.png.

Running the Similarity Analysis Script

python genetic_similarity_analysis.py

This will generate the following visualizations:

• Cosine Similarity Matrix: Cosine_Similarity_Matrix.png
• Euclidean Distance Matrix: Euclidean_Distance_Matrix.png
• Pearson Correlation Matrix: Pearson_Correlation_Matrix.png
• Spearman Correlation Matrix: Spearman_Correlation_Matrix.png

Output

Dendrogram for hierarchical clustering saved as gene_hierarchy_with_distances.png. Heatmaps for similarity and distance metrics saved as separate PNG files.