Abstract:Deep graph neural networks are extensively utilized to predict chemical reactivity and molecular properties. However, because of the complexity of chemical space, such models often have difficulty extrapolating beyond the chemistry contained in the training set. Augmented model with quantum mechanical (QM) descriptors is anticipated to improve its generalizability. However, obtaining QM descriptors often requires CPU-intensive computational chemistry calculations. To identify when QM descriptors help graph neural networks predict chemical properties, we conduct a systematic investigation of the impact of atom, bond, and molecular QM descriptors on the performance of directed message passing neural networks (D-MPNNs) for predicting 16 molecular properties. The analysis surveys computational and experimental targets, classification and regression tasks, and varied dataset sizes from several hundred to hundreds of thousands of datapoints. Our results indicate that QM descriptors are mostly beneficial for D-MPNN performance on small datasets, provided that the descriptors correlate well with the targets and can be readily computed with high accuracy. Otherwise, using QM descriptors can add cost without benefit or even introduce unwanted noise that can degrade model performance. Strategic integration of QM descriptors with D-MPNN unlocks potential for physics-informed, data-efficient modeling with some interpretability that can streamline de novo drug and material designs. To facilitate the use of QM descriptors in machine learning workflows for chemistry, we provide a set of guidelines regarding when and how to best leverage QM descriptors, a high-throughput workflow to compute them, and an enhancement to Chemprop, a widely adopted open-source D-MPNN implementation for chemical property prediction.

What problem does this paper attempt to address?

The paper primarily explores the role of Quantum Mechanical (QM) descriptors in predicting chemical properties using Graph Neural Networks (GNNs) and their optimal application conditions. Specifically, the paper aims to address the following key questions: 1. **Effectiveness of QM Descriptors**: The paper aims to systematically study the impact of atomic, bond, and molecular-level QM descriptors on the performance of Directed Message Passing Neural Networks (D-MPNNs) in predicting 16 different molecular properties. 2. **Impact of Dataset Size**: The study examines the effect of different dataset sizes (ranging from a few hundred to several hundred thousand data points) on model performance and analyzes the differences in performance between classification and regression tasks. 3. **Best Practices for QM Descriptors**: Based on the research findings, the paper proposes a decision flowchart to guide how and when to best utilize QM descriptors to enhance the performance of GNN models. 4. **Computation and Integration of QM Descriptors**: The paper provides a high-throughput workflow for computing these descriptors and improves the Chemprop software package to support the efficient integration of QM descriptors. The core objective of the paper is to determine when QM descriptors help improve the accuracy of GNNs in predicting chemical properties, especially in data-limited scenarios. The study finds that QM descriptors are generally beneficial for smaller datasets, provided that these descriptors are highly relevant to the target properties and can be accurately computed. However, as the dataset size increases, the advantage of QM descriptors gradually diminishes. Additionally, the paper offers a series of recommendations on how to select and effectively utilize QM descriptors to optimize model performance and streamline the design process of new drugs and materials.

When Do Quantum Mechanical Descriptors Help Graph Neural Networks Predict Chemical Properties?

When Do Quantum Mechanical Descriptors Help Graph Neural Networks to Predict Chemical Properties?

Quantum chemistry-augmented neural networks for reactivity prediction: Performance, generalizability, and explainability

Molecular Property Prediction: A Multilevel Quantum Interactions Modeling Perspective

Universal neural network potentials as descriptors: Towards scalable chemical property prediction using quantum and classical computers

Toward Accurate Predictions of Atomic Properties via Quantum Mechanics Descriptors Augmented Graph Convolutional Neural Network: Application of This Novel Approach in NMR Chemical Shifts Predictions

Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models

Introducing Block Design in Graph Neural Networks for Molecular Properties Prediction

Hybrid Quantum Graph Neural Network for Molecular Property Prediction

Analyzing Learned Molecular Representations for Property Prediction

Analysis of Atom-level pretraining with Quantum Mechanics (QM) data for Graph Neural Networks Molecular property models

Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Properties

Quantum-chemical insights from deep tensor neural networks

Extreme Acceleration of Graph Neural Network-based Prediction Models for Quantum Chemistry

Transferable Multilevel Attention Neural Network for Accurate Prediction of Quantum Chemistry Properties via Multitask Learning

Flexible dual-branched message passing neural network for quantum mechanical property prediction with molecular conformation

Quantum Mechanics and Machine Learning Synergies: Graph Attention Neural Networks to Predict Chemical Reactivity

PointGAT: A quantum chemical property prediction model integrating graph attention and 3D geometry

Neural Message Passing for Quantum Chemistry

Directed graph attention neural network utilizing 3D coordinates for molecular property prediction