2024_Ruthe_SND/code/heat_maps_correlation.py

import os
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

# Set up the directory paths
code_dir = os.path.dirname(os.path.abspath(__file__))
parent_dir = os.path.dirname(code_dir)
csv_file = os.path.join(parent_dir, 'output', "Correlation_with_behavior", 'correlation_dti_with_behavior.csv')
output_dir = os.path.dirname(csv_file)

# Read the CSV file into a DataFrame
df = pd.read_csv(csv_file)
df = df[df["dialation_amount"] == 1]

# Get unique Qtypes
qtypes = df["Qtype"].unique()
qtypes = ["fa"]

# Iterate over unique values of 'Qtype'
for qtype in qtypes:
    # Filter the DataFrame for the current 'Qtype'
    filtered_df = df[df["Qtype"] == qtype]
    
    # Get unique groups within the filtered DataFrame
    groups = filtered_df["Group"].unique()
    
    # Set up figure for the current Qtype
    num_groups = len(groups)
    fig, axes = plt.subplots(num_groups, 2, figsize=(16, num_groups * 6))
    
    # Iterate over unique values of 'Group' within the current Qtype
    for i, group in enumerate(groups):
        # Filter the DataFrame for the current 'Group'
        group_df = filtered_df[filtered_df["Group"] == group]
        
        # Pivot the filtered DataFrame to prepare for heatmap
        for j, value in enumerate(['R', 'Pval']):
            heatmap_df = group_df.pivot(index='mask_name', columns='merged_timepoint', values=value)
            
            # Plot heatmap
            if value == "R":
                sns.heatmap(heatmap_df, ax=axes[i, j], cmap='coolwarm', annot=True, fmt=".2f", cbar=True)
            else:
                sns.heatmap(heatmap_df, ax=axes[i, j], cmap='coolwarm_r', annot=True, fmt=".2f", cbar=True)
            
            axes[i, j].set_title(f'Group: {group}, Qtype: {qtype}, Value: {value}')
            axes[i, j].set_xlabel('Time Point', fontsize=8)
            axes[i, j].set_ylabel('Tract Name', fontsize=8)
            axes[i, j].tick_params(axis='both', which='major', labelsize=8)
    
    plt.tight_layout()
    save_path = os.path.join(output_dir, f'heatmap_{qtype}_2dialation_corrected_spearman.png')
    #plt.savefig(save_path, dpi=300)
    plt.show()