suplfig4.py

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

# Load data for the other plots
CSmethod = "UCBShift"
data_source_file = './csp_stats_' + CSmethod + '.csv'
df = pd.read_csv(data_source_file)

# Setting up the figure and subplots
fig, axs = plt.subplots(2, 2, figsize=(20, 20), gridspec_kw={'height_ratios': [1, 1], 'width_ratios': [1, 1]})
plt.subplots_adjust(hspace=0.4, wspace=0.4)  # Increase spacing between subplots

# Calculate residuals for each pair
df['residual_2'] = df['F1_AF2'] - df['F1']

sns.scatterplot(ax=axs[0, 0], x=df['F1'], y=df['F1_AF2'])
axs[0, 0].plot([0, 1], [0, 1], color='red')  # y=x line
axs[0, 0].set_title('NMR vs AF2 F1 Scores', fontsize=20)
axs[0, 0].set_xlabel('PDB F1 Score', fontsize=20)
axs[0, 0].set_ylabel('AF2 F1 Score', fontsize=20)
axs[0, 0].text(0.7, 0.1, 'PDB>AF2', color='red', fontweight='bold', fontsize=16, verticalalignment='bottom', horizontalalignment='left', transform=axs[0, 0].transAxes)
axs[0, 0].text(0.1, 0.9, 'AF2>PDB', color='red', fontweight='bold', fontsize=16, verticalalignment='top', horizontalalignment='left', transform=axs[0, 0].transAxes)

# Histogram of residuals for consensus_AF2
data = df['residual_2']
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[0, 1], data=data, kde=True, bins=bins)
axs[0, 1].set_title('Histogram of Residuals for F1 Scores', fontsize=20)
axs[0, 1].set_xlabel('Residual', fontsize=20)
axs[0, 1].set_ylabel('Frequency', fontsize=20)
axs[0, 1].axvline(x=0.0, color='red', linestyle='-', linewidth=2)
axs[0, 1].text(0.7, 0.7, 'AF2>PDB', color='red', fontweight='bold', fontsize=16, verticalalignment='bottom', horizontalalignment='left', transform=axs[0, 1].transAxes)
axs[0, 1].text(0.05, 0.7, 'PDB>AF2', color='red', fontweight='bold', fontsize=16, verticalalignment='top', horizontalalignment='left', transform=axs[0, 1].transAxes)
# Load data for the other plots

# Calculate residuals for each pair
df['residual_2'] = df['MCC_AF2'] - df['MCC']

sns.scatterplot(ax=axs[1, 0], x=df['MCC'], y=df['MCC_AF2'])
axs[1, 0].plot([-1, 1], [-1, 1], color='red')  # y=x line
axs[1, 0].set_title('NMR vs AF2 MCCs', fontsize=20)
axs[1, 0].set_xlabel('PDB MCCs', fontsize=20)
axs[1, 0].set_ylabel('AF2 MCCs', fontsize=20)
axs[1,0].set_xlim((-1,1))
axs[1,0].set_ylim((-1,1))
axs[1, 0].text(0.7, 0.1, 'PDB>AF2', color='red', fontweight='bold', fontsize=16, verticalalignment='bottom', horizontalalignment='left', transform=axs[1, 0].transAxes)
axs[1, 0].text(0.1, 0.9, 'AF2>PDB', color='red', fontweight='bold', fontsize=16, verticalalignment='top', horizontalalignment='left', transform=axs[1, 0].transAxes)

# Histogram of residuals for consensus_AF2
data = df['residual_2']
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[1, 1], data=data, kde=True, bins=bins)
axs[1, 1].set_title('Histogram of Residuals for MCCs', fontsize=20)
axs[1, 1].set_xlabel('Residual', fontsize=20)
axs[1, 1].set_ylabel('Frequency', fontsize=20)
axs[1, 1].axvline(x=0.0, color='red', linestyle='-', linewidth=2)
axs[1, 1].text(0.7, 0.7, 'AF2>PDB', color='red', fontweight='bold', fontsize=16, verticalalignment='bottom', horizontalalignment='left', transform=axs[1, 1].transAxes)
axs[1, 1].text(0.05, 0.7, 'PDB>AF2', color='red', fontweight='bold', fontsize=16, verticalalignment='top', horizontalalignment='left', transform=axs[1, 1].transAxes)

if False:
    # Calculate residuals for each pair
    df['residual_2'] = df['consensus_AF2'] - df['consensus']

    sns.scatterplot(ax=axs[2, 0], x=df['consensus'], y=df['consensus_AF2'])
    axs[2, 0].plot([0, 1], [0, 1], color='red')  # y=x line
    axs[2, 0].set_title('NMR vs AF2 CSP_Rank_Scores', fontsize=20)
    axs[2, 0].set_xlabel('PDB CSP_Rank_Score', fontsize=20)
    axs[2, 0].set_ylabel('AF2 CSP_Rank_Score', fontsize=20)
    axs[2, 0].text(0.7, 0.1, 'PDB>AF2', color='red', fontweight='bold', fontsize=16, verticalalignment='bottom', horizontalalignment='left', transform=axs[2, 0].transAxes)
    axs[2, 0].text(0.1, 0.9, 'AF2>PDB', color='red', fontweight='bold', fontsize=16, verticalalignment='top', horizontalalignment='left', transform=axs[2, 0].transAxes)

    # Histogram of residuals for consensus_AF2
    data = df['residual_2']
    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[2, 1], data=data, kde=True, bins=bins)
    axs[2, 1].set_title('Histogram of Residuals for CSP_Rank_Score', fontsize=20)
    axs[2, 1].set_xlabel('Residual', fontsize=20)
    axs[2, 1].set_ylabel('Frequency', fontsize=20)
    axs[2, 1].axvline(x=0.0, color='red', linestyle='-', linewidth=2)
    axs[2, 1].text(0.7, 0.7, 'AF2>PDB', color='red', fontweight='bold', fontsize=16, verticalalignment='bottom', horizontalalignment='left', transform=axs[2, 1].transAxes)
    axs[2, 1].text(0.05, 0.7, 'PDB>AF2', color='red', fontweight='bold', fontsize=16, verticalalignment='top', horizontalalignment='left', transform=axs[2, 1].transAxes)

# Adjust spacing between subplots
plt.subplots_adjust(hspace=0.45, wspace=0.15)

plt.show()