plot_PSVS_richardson.py

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

# Load the updated CSV file
df = pd.read_csv('./csp_stats.csv')

# List of columns to plot for histograms
columns_to_plot_af2 = ["AF2_Most_favoured", "AF2_Allowed", "AF2_Disallowed"]
columns_to_plot_nmr = ["NMR_Most_favoured", "NMR_Allowed", "NMR_Disallowed"]
suffix = '_trim'

fontsize_t = 16
# List of columns to plot for histograms
columns_to_plot = columns_to_plot_nmr + columns_to_plot_af2

# Setting up the figure and subplots in a 2x2 grid
fig, axs = plt.subplots(3, 3, figsize=(10, 10))

# Flatten the axs array for easier indexing
axs = axs.flatten()

for idx, column in enumerate(columns_to_plot):
    axs[idx].hist(df[column+suffix].dropna(), bins=15, color='blue', alpha=0.7)
    #axs[idx].set_title(f'Histogram of {column}', fontsize=14)
    axs[idx].set_xlabel(column.replace('NMR','PDB').replace('_',' '), fontsize=fontsize_t)
    axs[idx].set_ylabel('Frequency', fontsize=fontsize_t)
    if 'Disallowed' in column:
        axs[idx].set_xlim(0,12)
    elif 'Allowed' in column:
        axs[idx].set_xlim(0,30)
    elif 'favoured' in column:
        axs[idx].set_xlim(65,100)
    #axs[idx].set_xlim(-10, 3)

# Creating scatter plots comparing AF2 and NMR data
for idx, (nmr_col, af2_col) in enumerate(zip(columns_to_plot_nmr, columns_to_plot_af2)):
    scatter_idx = idx + 6  # Scatter plots will be the last two plots in the grid
    sns.scatterplot(ax=axs[scatter_idx], x=df[nmr_col+suffix], y=df[af2_col+suffix])
    #axs[scatter_idx].set_title(f'Scatter plot of {af2_col} vs {nmr_col}', fontsize=14)
    axs[scatter_idx].set_xlabel(nmr_col.replace('NMR', 'PDB').replace('_', ' '), fontsize=fontsize_t)
    axs[scatter_idx].set_ylabel(af2_col.replace('_',' '), fontsize=fontsize_t)
    axs[scatter_idx].plot([df[nmr_col+suffix].min(), df[nmr_col+suffix].max()], [df[nmr_col+suffix].min(), df[nmr_col+suffix].max()], color='red', linestyle='--')  # y=x line for reference
    if 'Disallowed' in nmr_col:
        axs[scatter_idx].set_xlim(0,12)
        axs[scatter_idx].set_ylim(0,12)
    elif 'Allowed' in nmr_col:
        axs[scatter_idx].set_xlim(0,30)
        axs[scatter_idx].set_ylim(0,30)
    elif 'favoured' in nmr_col:
        axs[scatter_idx].set_xlim(65,100)
        axs[scatter_idx].set_ylim(65,100)

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