Appendix E. Extended Technical Material

This appendix contains technical deep dives, extended derivations, and complete implementations that complement the print edition chapters. Each section corresponds to a specific chapter and is referenced at the relevant point in the print text via "TECHNICAL DEEP DIVE" callout boxes.

E.1 Extended Regularization and Hyperparameter Tuning (Ch 3)

This appendix extends the hERG channel blockade classification pipeline developed in Chapter 3. It provides the full treatment of foundational ML concepts (data generating distributions, the learning algorithm, and parametric versus nonparametric models) that were summarized in Section 3.3.1, as well as the detailed regularization analysis and hyperparameter tuning walkthrough condensed in Sections 3.4.1 and 3.4.3. Readers who want a deeper understanding of the theoretical underpinnings and the complete Scikit-Learn pipeline configurations used to arrive at Chapter 3's final model should work through this material alongside the chapter.

E.1.1 Learning from Data

We can formally define our problem as a mathematical function, f∶ X→Y, that maps the set of all possible inputs, X, to the set of all possible outputs, Y. In this case, the set of all possible inputs is the full chemical space of druglike compounds, and the set of all possible outputs is a simple "yes" or "no" determination about whether an input compound will be hERG blocking.

E.1.2 Detailed Regularization Analysis

E.2 Gradient Descent Derivation and Scikit-Learn Linear Model Tour (Chapter 4)

E.2.1 OLS Derivation and Gradient Descent Walkthrough

E.2.2 Extended Scikit-Learn Linear Model Tour

E.3 Deep Docking Complete Implementation (Chapter 9)

E.3.1 Prepare the Protein & Ligand Structures

E.3.2 Docking Validation & Interaction Fingerprints

E.4 Creating a Continuous Chemical Universe: Variational Autoencoders (Chapter 10)

E.4.1 The Variational Autoencoder

E.4.2 Posterior Collapse and Cyclic VAE

E.4.3 Monitoring Metrics

E.4.4 Training & Evaluating VAE-CYC

E.5 Graph Neural Network Foundations (Chapter 11)

E.5.1 The Engine of GNNs: Message Passing

E.5.2 Graph Convolutional Networks

E.5.3 Graph Attention Networks (GATs)

E.5.4 Graph Isomorphism Networks (GINs)

E.5.5 Graph-level Pooling for Molecular Representations

E.5.6 Challenges in Deep GNNs

E.6 SimpleFold: An End-to-End Example (Chapter 12)

E.6.1 Stage 1: Configuration

E.6.2 Stage 2: Loading Pretrained Models

E.6.3 Stage 3: Inference Pipeline

E.6.4 Stage 4: Flow Matching

E.6.5 Stage 5: From Sequence to Structure

E.6.6 Comparing Approaches: PLM-Based vs. MSA-Based

E.6.7 Retrospective