AI for Drug Discovery

Molecular Representation

SMILES and Molecular Graphs

from rdkit import Chem

def smiles_to_graph(smiles):
    mol = Chem.MolFromSmiles(smiles)
    atoms = [{'atomic_num': a.GetAtomicNum(), 'formal_charge': a.GetFormalCharge(),
              'hybridization': str(a.GetHybridization())} for a in mol.GetAtoms()]
    bonds = [{'begin': b.GetBeginAtomIdx(), 'end': b.GetEndAtomIdx(),
              'type': str(b.GetBondType())} for b in mol.GetBonds()]
    return {'atoms': atoms, 'bonds': bonds}

Molecular Fingerprints

Molecular Property Prediction

Graph Neural Networks

import torch
from torch_geometric.nn import GCNConv, global_mean_pool

class MolecularPropertyPredictor(torch.nn.Module):
    def __init__(self, in_channels, hidden_channels, out_channels):
        super().__init__()
        self.conv1 = GCNConv(in_channels, hidden_channels)
        self.conv2 = GCNConv(hidden_channels, hidden_channels)
        self.conv3 = GCNConv(hidden_channels, hidden_channels)
        self.fc = torch.nn.Linear(hidden_channels, out_channels)

    def forward(self, x, edge_index, batch):
        x = torch.relu(self.conv1(x, edge_index))
        x = torch.relu(self.conv2(x, edge_index))
        x = torch.relu(self.conv3(x, edge_index))
        x = global_mean_pool(x, batch)
        return self.fc(x)

Generative Models for Drug Design

Variational Autoencoders

Reinforcement Learning

class MolGenerator(nn.Module):
    def __init__(self, vocab_size, embed_dim, hidden_dim):
        super().__init__()
        self.embedding = nn.Embedding(vocab_size, embed_dim)
        self.lstm = nn.LSTM(embed_dim, hidden_dim, batch_first=True)
        self.output = nn.Linear(hidden_dim, vocab_size)

    def forward(self, x, hidden=None):
        embedded = self.embedding(x)
        output, hidden = self.lstm(embedded, hidden)
        return self.output(output), hidden

Virtual Screening Pipelines

Pharmacokinetic Properties

• Lipinski Rule of Five: MW < 500, LogP < 5, HBD <= 5, HBA <= 10
• Half-life:

Docking Score