CheMixHub: Datasets and Benchmarks for Chemical Mixture Property Prediction

2025-06-13

AI Summary

The lack of systematic benchmarks and standardized evaluation protocols hinders rigorous assessment of machine learning models for predicting properties of chemical mixtures. Method: This work introduces CheMixHub—the first comprehensive benchmark platform for mixture property prediction—encompassing 11 real-world industrial tasks (e.g., drug delivery, electrolyte design) and ~500K high-quality data points. It proposes a novel three-level data partitioning scheme—by component, formulation, and task—to rigorously evaluate contextual generalization. A unified baseline framework is developed, integrating graph neural networks with set-based representations, alongside a standardized evaluation protocol. Contribution/Results: CheMixHub fills a critical gap in ML-driven modeling of multi-molecular systems, significantly improving model comparability and robustness assessment. All data and code are publicly released to foster community-wide standardization and reproducible research.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Constraint Satisfaction and Optimization: Mixed Discrete/Continuous Optimization
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Developing improved predictive models for multi-molecular systems is crucial, as nearly every chemical product used results from a mixture of chemicals. While being a vital part of the industry pipeline, the chemical mixture space remains relatively unexplored by the Machine Learning community. In this paper, we introduce CheMixHub, a holistic benchmark for molecular mixtures, covering a corpus of 11 chemical mixtures property prediction tasks, from drug delivery formulations to battery electrolytes, totalling approximately 500k data points gathered and curated from 7 publicly available datasets. CheMixHub introduces various data splitting techniques to assess context-specific generalization and model robustness, providing a foundation for the development of predictive models for chemical mixture properties. Furthermore, we map out the modelling space of deep learning models for chemical mixtures, establishing initial benchmarks for the community. This dataset has the potential to accelerate chemical mixture development, encompassing reformulation, optimization, and discovery. The dataset and code for the benchmarks can be found at: https://github.com/chemcognition-lab/chemixhub

Problem

• Predict properties of chemical mixtures using ML models
• Address lack of benchmarks for multi-molecular system prediction
• Enable reformulation and discovery in industrial chemical applications

Innovation

• Introduces CheMixHub benchmark for molecular mixtures
• Includes 500k data points from 7 datasets
• Assesses model robustness with data splitting