Abstract:One of the major applications of generative models for drug discovery targets the lead-optimization phase. During the optimization of a lead series, it is common to have scaffold constraints imposed on the structure of the molecules designed. Without enforcing such constraints, the probability of generating molecules with the required scaffold is extremely low and hinders the practicality of generative models for de novo drug design. To tackle this issue, we introduce a new algorithm, named SAMOA (Scaffold Constrained Molecular Generation), to perform scaffold-constrained in silico molecular design. We build on the well-known SMILES-based Recurrent Neural Network (RNN) generative model, with a modified sampling procedure to achieve scaffold-constrained generation. We directly benefit from the associated reinforcement learning methods, allowing to design molecules optimized for different properties while exploring only the relevant chemical space. We showcase the method's ability to perform scaffold-constrained generation on various tasks: designing novel molecules around scaffolds extracted from SureChEMBL chemical series, generating novel active molecules on the Dopamine Receptor D2 (DRD2) target, and finally, designing predicted actives on the MMP-12 series, an industrial lead-optimization project.The Supporting Information is available free of charge at <a class="ext-link" href="/doi/10.1021/acs.jcim.0c01015?goto=supporting-info">https://pubs.acs.org/doi/10.1021/acs.jcim.0c01015</a>.Availability of data used to train models and run experiments, algorithmic description of naive policy masking (Algorithm S1), algorithmic description of hill-climbing (Algorithm S2), examples of molecules obtained after sampling the core of fexofenadine (Figure S1 for sampled molecules, Figure S2 for the original molecule), examples of molecules obtained after replacing the ring system of fexofenadine (Figure S3), distributions of ClogP, Molecular Weight, QED and SAS for reference molecules, molecules where a branched decoration was sampled, and molecules where the core was sampled (Figure S4), the 17 validation scaffolds from SureChEMBL (Figure S5), histograms of Tanimoto similarities between generated molecules for each of the DRD2 scaffolds (Figure S6), and unicity of sampled decorations for each open positions of the DRD2 scaffolds (Figure S7) (<a class="ext-link" href="/doi/suppl/10.1021/acs.jcim.0c01015/suppl_file/ci0c01015_si_001.pdf">PDF</a>)This article has not yet been cited by other publications.

Microphase separation induced by interfacial segregation of isotropic, spherical nanoparticles.

RASA: A Rapid Retrosynthesis-Based Scoring Method for the Assessment of Synthetic Accessibility of Drug-like Molecules

Estimating the synthetic accessibility of molecules with building block and reaction-aware SAScore

DFRscore: Deep Learning-Based Scoring of Synthetic Complexity with Drug-Focused Retrosynthetic Analysis for High-Throughput Virtual Screening

FSscore: A Machine Learning-based Synthetic Feasibility Score Leveraging Human Expertise

Generate What You Can Make: Achieving in-house synthesizability with readily available resources in de novo drug design

SDDBench: A Benchmark for Synthesizable Drug Design

Directly Optimizing for Synthesizability in Generative Molecular Design using Retrosynthesis Models

RetroGNN: Fast Estimation of Synthesizability for Virtual Screening and De Novo Design by Learning from Slow Retrosynthesis Software

DrugSynthMC: an atom based generation of drug-like molecules with Monte Carlo Search

Organic Compound Synthetic Accessibility Prediction Based on the Graph Attention Mechanism

SynthFormer: Equivariant Pharmacophore-based Generation of Molecules for Ligand-Based Drug Design

MolScore: a scoring, evaluation and benchmarking framework for generative models in de novo drug design

ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generation

Leap: molecular synthesisability scoring with intermediates

Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning

Assessing Methods and Obstacles in Chemical Space Exploration

Scaffold-Constrained Molecular Generation

MolScore: A scoring and evaluation framework for de novo drug design

Path Toward High-Throughput Synthesis Planning via Performance Benchmarking

Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learning