Navigating the Fragrance space Via Graph Generative Models And Predicting Odors
This work addresses the challenge of efficient fragrance discovery for researchers and industry, though it is incremental as it applies existing generative and predictive methods to a specific domain.
The paper tackled the problem of navigating chemical space for fragrance discovery by generating molecules and predicting their odor likeliness, achieving a ROC AUC score of 0.97 for odor prediction.
We explore a suite of generative modelling techniques to efficiently navigate and explore the complex landscapes of odor and the broader chemical space. Unlike traditional approaches, we not only generate molecules but also predict the odor likeliness with ROC AUC score of 0.97 and assign probable odor labels. We correlate odor likeliness with physicochemical features of molecules using machine learning techniques and leverage SHAP (SHapley Additive exPlanations) to demonstrate the interpretability of the function. The whole process involves four key stages: molecule generation, stringent sanitization checks for molecular validity, fragrance likeliness screening and odor prediction of the generated molecules. By making our code and trained models publicly accessible, we aim to facilitate broader adoption of our research across applications in fragrance discovery and olfactory research.