GraphNOSE: A Graph Transformer in Olfaction

Researchers introduced GraphNOSE, an open-source graph transformer framework that predicts multi-label odor descriptors from SMILES strings for single molecules and binary mixtures, according to an arXiv paper (2609.05694v1). GraphNOSE outperforms linear models, molecular language model embeddings, molecular fingerprints, and baseline graph neural networks by an average AUROC margin of 4.52% (p < 0.01) while using six times fewer parameters than standard GNN baselines, and reaches 84% AUROC on out-of-distribution compounds versus 81% for the prior state-of-the-art GNN Open-POM (p < 0.001). The authors also applied explainable AI methods to identify substructures and molecular features driving odor predictions.

arXiv:2609.05694v1 Announce Type: new Abstract: Predicting olfactory qualities from molecular structure is an open problem in chemoinformatics. Although linear models can link molecular features to odor descriptors, they often fail when extrapolating to novel chemical scaffolds, extreme molecular weights, or complex odor mixtures. To address this, we introduce GraphNOSE, an open-source graph transformer framework that predicts multi-label odor descriptors from simplified molecular-input line-entry system SMILES strings for single molecules and binary mixtures. By integrating positional and structural encodings within a transformer-based graph architecture, GraphNOSE achieves strong performance with six times fewer parameters than standard graph neural network GNN baseline while consistently outperforming linear models, molecular language model embeddings, molecular fingerprints, and baseline GNNs by an average area under the ROC curve AUROC margin of 4.52% p < 0.01 . GraphNOSE achieves an AUROC of 84% on out-of-distribution compounds OODs . This exceeds the current state-of-the-art GNN for OOD in olfaction Open-POM: 81%, p < 0.001 , and identifies conditions under which linear models empirically fail. Finally, we apply XAI explainable AI methods to identify which substructures and molecular features drive odor predictions, yielding insights consistent with chemical intuition and grounded in the model's learned representations. Together, these results establish GraphNOSE as a scalable and interpretable architecture for olfactory prediction that generalizes to structurally distinct compounds underrepresented in current perceptual databases.