Search papers, labs, and topics across Lattice.
This paper introduces a chemistry-guided classical machine learning branch, utilizing Morgan fingerprints, RDKit descriptors, and conformer-derived geometric features, to complement a CoMPT graph predictor. By fusing the out-of-fold predictions of four tree learners from the classical branch with the graph predictor via Ridge regression, the approach achieves state-of-the-art accuracy on molecular property regression tasks like ESOL, FreeSolv, and Lipophilicity. The findings demonstrate that simple prediction-level fusion of chemically informed classical models and graph neural networks can outperform complex multimodal architectures.
Simple fusion of chemically-guided classical models with graph predictors unlocks competitive molecular property regression, challenging the necessity of complex multimodal architectures.
Accurate prediction of continuous molecular properties supports chemical discovery and design. Many recent approaches combine multiple molecular representations within increasingly complex neural architectures, aiming to achieve higher accuracy. However, classical machine-learning models remain competitive when trained on chemically meaningful molecular representations. We therefore constructed a chemistry-guided classical branch from Morgan fingerprints, RDKit molecular descriptors, and conformer-derived geometric features selected to provide complementary views of molecular structure. We combined four tree learners trained on these representations with a CoMPT graph predictor and integrated their out-of-fold predictions using Ridge regression. Across ESOL, FreeSolv, and Lipophilicity datasets, the fused model achieved lower mean root-mean-square error (RMSE) than Chemprop and competitive graph and multimodal methods, including CoMPT, MoleSG, KFLM2, and MCMPP. After retraining on the separately curated AqSolDB dataset, the same approach, without redesign or retuning, reduced mean absolute error (MAE) relative to the classical ensemble and achieved lower mean MAE than Chemprop. These results show that accurate and transferable molecular property regression does not necessarily require a complex joint multimodal architecture: classical learning remains competitive when chemical insight guides the choice of molecular representations, and a graph model can add complementary information through simple prediction-level fusion.
