Abstract:Deep generative models yielding transition metal complexes (TMCs) remain scarce despite the key role of these compounds in industrial catalytic processes, anticancer therapies, and energy transformations. Compared to drug discovery within the organic molecular space, TMCs pose further challenges including the encoding of chemical bonds of higher complexity and the optimization of multiple properties, in a context in which synthesizability is affected by additional, complex factors. In this work, we developed a junction tree variational autoencoder (JT-VAE) model for the generation of metal ligands. After implementing a SMILES-based encoding of the metal–ligand bonds, the model was trained with the tmQMg-L ligand library, allowing for the random generation of thousands of monodentate and bidentate ligands with full validity and high novelty. The generated ligands were labeled with two target properties of the associated [IrL4]+ and [IrL2]+ homoleptic TMCs; namely the HOMO-LUMO gap (ϵ) and the metal charge (qIr), both computed at a DFT level. This data was used to implement a conditional JT-VAE model generating ligands from a prompt, with the single or dual objective of optimizing either one or both properties in Y = (ϵ, qIr). Conditional ligand generation was able to navigate both central and extreme regions of this bidimensional property space, allowing for chemical interpretation based on the step-wise analysis of the decoded optimization trajectories.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to achieve the dual - objective inverse design of transition - metal complexes (TMCs) through deep generative models. Specifically, the goals of the paper are as follows: 1. **Develop a deep generative model for generating metal ligands**: Existing deep generative models are less applied in generating transition - metal complexes, especially in generating ligands with specific properties. The paper develops a model based on Junction - Tree Variational Autoencoder (JT - VAE), which can generate monodentate and polydentate metal ligands, and these ligands have high validity and novelty. 2. **An explicit method for representing metal coordination**: Different from organic - molecule - generation models, models for generating TMC ligands need to explicitly represent metal - ligand bonds. The paper proposes a method based on SMILES strings to represent metal - ligand bonds, thus ensuring that the generated ligands have the correct coordination patterns. 3. **Conditional generative model**: In addition to unconditionally generating ligands, the paper also develops a conditional generative model, which can generate ligands according to one or two target properties (such as HOMO - LUMO energy gap \( \epsilon \) and the local charge of iridium \( q_{\text{Ir}} \)). These properties are crucial for the stability and reactivity of TMCs. 4. **Optimize the generated ligands**: By training the model, it can generate the optimal ligands according to the target properties. The paper uses DFT calculations to verify the properties of the generated TMCs and optimizes the target properties by the gradient - descent method. 5. **Evaluate the quality of the generated ligands**: The paper evaluates the quality of the generated ligands from multiple perspectives, including chemical validity, uniqueness, novelty, diversity, and synthetic feasibility. Through these evaluations, it is ensured that the generated ligands are not only theoretically reasonable but also have the possibility of synthesis in experiments. In conclusion, this paper aims to solve the inverse - design problem of transition - metal complexes through deep generative models, especially in generating ligands with specific properties, and provides new tools and methods for fields such as catalysis, anti - cancer treatment, and energy conversion.

Deep Generative Model for the Dual-Objective Inverse Design of Metal Complexes

Generative Design of Functional Metal Complexes Utilizing the Internal Knowledge of Large Language Models

Inverse design of ligands using a deep generative model semi‐supervised by a data‐driven ligand field strength metric

PrefixMol: Target- and Chemistry-aware Molecule Design Via Prefix Embedding

Exploiting Ligand Additivity for Transferable Machine Learning of Multireference Character Across Known Transition Metal Complex Ligands

Deep generative models for ligand‐based de novo design applied to multi‐parametric optimization

Partial to Total Generation of 3D Transition Metal Complexes

Modelling ligand exchange in metal complexes with machine learning potentials

Generative Design of inorganic compounds using deep diffusion language models

Graph neural networks for predicting metal–ligand coordination of transition metal complexes

Generative Organic Electronic Molecular Design Informed by Quantum Chemistry

LigandDiff: 3D Transition Metal Complex Generation with Diffusion Models

Enabling Inverse Design in Chemical Compound Space: Mapping Quantum Properties to Structures for Small Organic Molecules

Many-body Expansion Based Machine Learning Models for Octahedral Transition Metal Complexes

Generative design of compounds with desired potency from target protein sequences using a multimodal biochemical language model

Pathway-Guided Optimization of Deep Generative Molecular Design Models for Cancer Therapy

Generative model based on junction tree variational autoencoder for HOMO value prediction and molecular optimization

Generative Chemical Transformer: Neural Machine Learning of Molecular Geometric Structures from Chemical Language via Attention

Automated design of multi-target ligands by generative deep learning

Ligand Many-Body Expansion as a General Approach for Accelerating Transition Metal Complex Discovery