suplfig1.py

import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
import numpy as np

# Define the list of methods
CSmethods = [ 'UCBShift', 'SPARTA','ShiftX']

# Setting up the figure and subplots
fig, axs = plt.subplots(len(CSmethods), 2, figsize=(10, 3 * len(CSmethods)), gridspec_kw={'height_ratios': [1] * len(CSmethods), 'width_ratios': [1, 1]})

for idx, CSmethod in enumerate(CSmethods):
    data_source_file = f'./csp_stats_{CSmethod}.csv'
    df = pd.read_csv(data_source_file)

    # Calculate residuals for each pair
    df['residual_1'] = df['F1_AF2'] - df['F1']
    df['residual_2'] = df['MCC_AF2'] - df['MCC']
    df['residual_3'] = df['consensus_AF2'] - df['consensus']

    # Scatter plot for consensus vs consensus_AF2
    sns.scatterplot(ax=axs[idx, 0], x=df['consensus'], y=df['consensus_AF2'])
    axs[idx, 0].plot([0, 1], [0, 1], color='red')  # y=x line
    axs[idx, 0].set_title(f'PDB vs AF2 {CSmethod} CSP_Rank_Scores', fontsize=14)
    axs[idx, 0].set_xlabel('PDB Score', fontsize=12)
    axs[idx, 0].set_ylabel('AF2 Score', fontsize=12)
    axs[idx, 0].text(0.7, 0.1, 'PDB>AF2', color='red', fontweight='bold', fontsize=12, verticalalignment='bottom', horizontalalignment='left', transform=axs[idx, 0].transAxes)
    axs[idx, 0].text(0.1, 0.9, 'AF2>PDB', color='red', fontweight='bold', fontsize=12, verticalalignment='top', horizontalalignment='left', transform=axs[idx, 0].transAxes)

    # Histogram of residuals for consensus_AF2
    data = df['residual_3']
    max_abs_value = max(abs(data.min()), abs(data.max()))
    symmetric_range = (-max_abs_value, max_abs_value)
    bin_width = (symmetric_range[1] - symmetric_range[0]) / 20
    bins = np.arange(symmetric_range[0], symmetric_range[1] + bin_width, bin_width)

    sns.histplot(ax=axs[idx, 1], data=data, kde=True, bins=bins)
    axs[idx, 1].set_title(f'Histogram of Residuals for {CSmethod} CSP_Rank_Score', fontsize=14)
    axs[idx, 1].set_xlabel('Residual', fontsize=12)
    axs[idx, 1].set_ylabel('Frequency', fontsize=12)
    axs[idx, 1].axvline(x=0.0, color='red', linestyle='-', linewidth=2)
    axs[idx, 1].text(0.7, 0.7, 'AF2>PDB', color='red', fontweight='bold', fontsize=12, verticalalignment='bottom', horizontalalignment='left', transform=axs[idx, 1].transAxes)
    axs[idx, 1].text(0.05, 0.7, 'PDB>AF2', color='red', fontweight='bold', fontsize=12, verticalalignment='top', horizontalalignment='left', transform=axs[idx, 1].transAxes)

# Adjust the layout to add more space between the plots
plt.subplots_adjust(hspace=0.5)
plt.tight_layout()
plt.show()