Aug 2026· Journal of Chemical Physics· Vol 165 7· 0 citations· 52 references
Medicine
Abstract
Melting point (MP) is an important thermophysical property for the chemical process industry, yet accurate prediction of MP for organic compounds in the absence of experimental data remains challenging due to the complex interplay between molecular packing, intermolecular interactions, and electronic structure. Traditional group contribution and quantitative structure-property relationship models, which rely primarily on static molecular descriptors, often fail to capture these critical condensed-phase effects. In this study, we present a hybrid machine learning framework that integrates cheminformatics descriptors with quantum chemical features and dynamic condensed-phase descriptors derived from molecular dynamics (MD) simulations. Using a curated subset of the DIPPR 801 database, multiple machine learning architectures, including light gradient boosting machine (LightGBM) and graph convolutional networks, were evaluated with feature sets of increasing physical fidelity. The best-performing model, based on LightGBM trained on Dragon descriptors augmented with MD and quantum chemical features, achieves a mean absolute error of 22.5 K, outperforming descriptor-only models and structure-based deep learning baselines. Shapley additive explanations interpretability analysis reveals that melting behavior is governed primarily by molecular topology, surface-area-weighted electronic descriptors, and condensed-phase interaction properties. In contrast, many isolated functional group and single molecule electronic descriptors contribute negligibly once these effects are accounted for. These results demonstrate that incorporating physics-informed, multi-scale descriptors enables more accurate and physically interpretable MP predictions.
The glass transition temperature (Tg) of polyimides is a critical parameter determining their processability and application performance. Traditional experimental methods for measuring Tg are time‐consuming and costly, while existing machine learning prediction models predominantly rely on manually defined molecular descriptors, which often fail to fully capture detailed molecular structural information, limiting their prediction accuracy and generalization capability. To address this, this study proposes a hybrid feature engineering strategy combining Morgan fingerprints and molecular descriptors to comprehensively represent the chemical structure of polyimides. Based on a dataset of 1257 polyimide samples from a public database, we systematically compared six feature selection methods and employed multiple mainstream machine learning algorithms for modeling. The results show that the CATB model performed best, achieving a coefficient of determination (R2) of 0.882 and a mean absolute error (MAE) of 17.34 °C on an independent test set, with fivefold cross‐validation further confirming the model's robustness. SHAP interpretability analysis revealed the significant influence of key features such as the number of rotatable bonds, ether bonds, and ether‐linked oxyethylene units on Tg, providing clear guidance for molecular design. External validation demonstrated the model's strong generalization ability. This study not only achieves high‐precision and robust Tg prediction but also highlights the importance of hybrid feature strategies in polymer property modeling, offering a data‐driven foundation for the rational design of polyimides.
Peishuai Xing, Xiaodong Guo, Yang Wang et al.· Molecular Informatics· 0 citations
This study presents a data-driven machine learning approach to predict the melting points of organic compounds, leveraging both 2D and 3D molecular descriptors and indicates that ML models can significantly improve melting-point predictions, providing a robust tool for the scientific community.
Md Kamruzzaman, Alexander Landera, N. Menon et al.· Journal of Cheminformatics· 1 citation
An explainable machine-learning framework was developed for dielectric constant prediction using 52,168 crystalline materials extracted from the Joint Automated Repository for Various Integrated Simulations (JARVIS-DFT) database, demonstrating the complementary roles of electronic structure and elemental chemistry.
D. Pundhir, Ashok Kumar· Applied Physics A· 0 citations
A design principle is proposed for advanced metal nitride HEDMs: prioritizing high nitrogen-to-metal ratios, light metal elements, and structures wherein nitrogen atoms are spatially separated by the metal matrix, which minimizes N-N bonds and favors dominant M-N bonding.
Yaozhong Liu, Huifang Du, Caimu Wang et al.· Chinese Physics B· 0 citations
Aggregation‐induced emission (AIE) has revolutionized the design of photoluminescent materials by enabling strong solid‐state emission from molecularly nonemissive compounds. However, rational prediction of AIE properties remains challenging because photophysical behavior depends not only on molecular structure but also on aggregate‐state packing and measurement conditions. This study develops a quantitative and interpretable machine learning (ML) framework for predicting experimentally reported emission energies of AIE‐active molecules using continuous physicochemical descriptors derived from molecular structures. A dataset of 590 AIE luminogens—including conjugated organics, donor–acceptor (D–A) systems, silicon‐containing luminogens, and transition‐metal complexes—was analyzed using Gaussian process regression (GPR) combined with SHapley Additive exPlanations (SHAP). The optimized descriptor‐based model achieved moderate predictive performance (test
R
2
= 0.58) and provided chemically interpretable structure–property trends. Feature attribution indicated that nitrogen‐ and sulfur‐containing motifs, electrotopological‐state descriptors, Burden–CAS–University of Texas (BCUT) descriptors, and stereodefined vinylene units are statistically associated with lower emission energies within the present dataset. Morgan fingerprint baseline models showed higher random‐split accuracy, whereas leave‐one‐cluster‐out validation revealed cluster‐dependent degradation for structurally separated regions. This work therefore provides an interpretable initial screening strategy for AIE luminogens while clarifying the need for future models incorporating measurement conditions, solid‐state structural descriptors, and electronic‐structure‐informed features.
