Abstract:In the past decade, Artificial Intelligence driven drug design and discovery has been a hot research topic, where an important branch is molecule generation by generative models, from GAN-based models and VAE-based models to the latest diffusion-based models. However, most existing models pursue only the basic properties like validity and uniqueness of the generated molecules, a few go further to explicitly optimize one single important molecular property (e.g. QED or PlogP), which makes most generated molecules little usefulness in practice. In this paper, we present a novel approach to generating molecules with desirable properties, which expands the diffusion model framework with multiple innovative designs. The novelty is two-fold. On the one hand, considering that the structures of molecules are complex and diverse, and molecular properties are usually determined by some substructures (e.g. pharmacophores), we propose to perform diffusion on two structural levels: molecules and molecular fragments respectively, with which a mixed Gaussian distribution is obtained for the reverse diffusion process. To get desirable molecular fragments, we develop a novel electronic effect based fragmentation method. On the other hand, we introduce two ways to explicitly optimize multiple molecular properties under the diffusion model framework. First, as potential drug molecules must be chemically valid, we optimize molecular validity by an energy-guidance function. Second, since potential drug molecules should be desirable in various properties, we employ a multi-objective mechanism to optimize multiple molecular properties simultaneously. Extensive experiments with two benchmark datasets QM9 and ZINC250k show that the molecules generated by our proposed method have better validity, uniqueness, novelty, Fréchet ChemNet Distance (FCD), QED, and PlogP than those generated by current SOTA models.

Design of fragrant molecules through the incorporation of rough sets into computer-aided molecular design

Design of formulated fragrant products using rough set machine learning and molecular design tools

Incorporating Machine Learning in Computer-Aided Molecular Design for Fragrance Molecules

Design of fragrance molecules using computer-aided molecular design with machine learning

Molecule Generation and Optimization for Efficient Fragrance Creation

Unlocking comprehensive molecular design across all scenarios with large language model and unordered chemical language

fragSMILES: a Chemical String Notation for Advanced Fragment and Chirality Representation

Bayesian molecular design with a chemical language model

Intelligent Consensus Predictions of the Retention Index of Flavor and Fragrance Compounds Using 2D Descriptors

Advancing Odor Classification Models Enhanced by Scientific Machine Learning and Mechanistic Model: Probabilistic Weight Assignment for Odor Intensity Prediction and Uncertainty Analysis for Robust Fragrance Classification

Molecular Design Method Using a Reversible Tree Representation of Chemical Compounds and Deep Reinforcement Learning

Gotta be SAFE: A New Framework for Molecular Design

Deep Geometry Handling and Fragment-wise Molecular 3D Graph Generation

Computer Aided Recipe Design: Optimization of Polydisperse Chemical Mixtures Using Molecular Descriptors

Molecular Design Based on Integer Programming and Splitting Data Sets by Hyperplanes

Generative Organic Electronic Molecular Design Informed by Quantum Chemistry

Diffusing on Two Levels and Optimizing for Multiple Properties: A Novel Approach to Generating Molecules with Desirable Properties

Autonomous, multiproperty-driven molecular discovery: From predictions to measurements and back

Generative AI-Driven Molecular Design: Combining Predictive Models and Reinforcement Learning for Tailored Molecule Generation

FragGen: Towards 3D Geometry Reliable Fragment-based Molecular Generation