File size: 4,713 Bytes
adecf03
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.decomposition import PCA
from sklearn.manifold import TSNE
from rdkit import Chem
from rdkit.Chem import AllChem



def load_and_process_data(file_path_train, file_path_screening, threshold_price=100, threshold_mwt=1000):
    # Load train dataset
    train = pd.read_excel(file_path_train)
    train['Canonical SMILES'] = train['Canonical SMILES'].apply(lambda x: Chem.MolToSmiles(Chem.MolFromSmiles(x)) if pd.notnull(x) else None)
    train.dropna(subset=['Canonical SMILES'], inplace=True)
    train_blue = train[train['Reactivity'] == 1]
    train_red = train[train['Reactivity'] == -1]
    
    # Load screening dataset
    screening = pd.read_excel(file_path_screening)
    # Apply both price and molecular weight thresholds
    screening_filtered = screening[(screening['Cheapest Unit Price (USD per g/mL)'] < threshold_price) & 
                                   (screening['MWt'] < threshold_mwt)].copy()
    screening_filtered['Canonical SMILES'] = screening_filtered['SMILES'].apply(lambda x: Chem.MolToSmiles(Chem.MolFromSmiles(x)) if pd.notnull(x) else None)
    screening_filtered.dropna(subset=['Canonical SMILES'], inplace=True)
    screening_grey = screening_filtered
    
    return train_blue, train_red, screening_grey


def generate_fingerprints_and_reduce_dimensions(dfs):
    combined_smiles = pd.concat([df['Canonical SMILES'] for df in dfs if not df.empty])
    fingerprints = [AllChem.GetMorganFingerprintAsBitVect(Chem.MolFromSmiles(smiles), radius=2, nBits=1024) for smiles in combined_smiles]
    fingerprints_array = np.array([list(fp) for fp in fingerprints])
    
    # PCA
    pca = PCA(n_components=2)
    pca_result = pca.fit_transform(fingerprints_array)
    
    # t-SNE with explicit parameter settings to avoid FutureWarnings
    tsne = TSNE(n_components=2, random_state=42, init='random', learning_rate='auto')
    tsne_result = tsne.fit_transform(fingerprints_array)
    
    return pca_result, tsne_result, combined_smiles.index

def plot_combined_data(pca_result, tsne_result, indices, colors, labels):
    plt.figure(figsize=(12, 6))
    
    # Adjust the color list to include the RGBA tuple for grey with 20% transparency
    colors = ['blue', 'red', (0.5, 0.5, 0.5, 0.02)]  # Adjusting the grey color for 2% transparency
    
    # Plot PCA results
    plt.subplot(1, 2, 1)
    # Ensure grey points are plotted first for them to be in the background
    for i in [2, 0, 1]:  # Reordering to plot grey first, followed by blue and red
        idx = indices[i]
        plt.scatter(pca_result[idx, 0], pca_result[idx, 1], c=colors[i], label=labels[i])
    plt.title('PCA of Molecular Fingerprints')
    plt.xlabel('PC1')
    plt.ylabel('PC2')
    plt.legend()

    # Plot t-SNE results
    plt.subplot(1, 2, 2)
    # Same reordering logic for t-SNE plot
    for i in [2, 0, 1]:  # Reordering to plot grey first, followed by blue and red
        idx = indices[i]
        plt.scatter(tsne_result[idx, 0], tsne_result[idx, 1], c=colors[i], label=labels[i])
    plt.title('t-SNE of Molecular Fingerprints')
    plt.xlabel('t-SNE 1')
    plt.ylabel('t-SNE 2')
    plt.legend()
    
    plt.tight_layout()
    plt.show()


def main(file_path_train, file_path_screening, price=100, mwt=1000):
    train_blue, train_red, screening_grey = load_and_process_data(file_path_train, file_path_screening, price, mwt)
    pca_result, tsne_result, _ = generate_fingerprints_and_reduce_dimensions([train_blue, train_red, screening_grey])
    
    # Calculate indices for coloring
    len_blue = len(train_blue)
    len_red = len(train_red)
    len_grey = len(screening_grey)
    indices = [range(len_blue), range(len_blue, len_blue + len_red), range(len_blue + len_red, len_blue + len_red + len_grey)]
    
    print(f"Number of blue points: {len_blue}")
    print(f"Number of red points: {len_red}")
    print(f"Number of grey points: {len_grey}")
    
    colors = ['blue', 'red', 'grey']  # Blue for Reactivity 1, Red for Reactivity -1, Grey for Screening
    labels = ['Train Reactivity 1', 'Train Reactivity -1', 'Screening']
    plot_combined_data(pca_result, tsne_result, indices, colors, labels)
    return screening_grey
    
file_name = "Inventory-2024-03-30.xlsx"    
processed_file_name = 'processed_' + file_name    

# Example usage
screening_df = main('SF2. EChem Reaction Screening Dataset.xlsx', processed_file_name,50, 500)
screening_df.to_excel(processed_file_name, index=False) 
print("Processing complete. The new file is saved as:", processed_file_name)
print(screening_df50_500)