Molecule Generation and Optimization for Efficient Fragrance Creation

Bruno C. L. Rodrigues,Vinicius V. Santana,Sandris Murins,Idelfonso B. R. Nogueira
2024-02-19
Abstract:This research introduces a Machine Learning-centric approach to replicate olfactory experiences, validated through experimental quantification of perfume perception. Key contributions encompass a hybrid model connecting perfume molecular structure to human olfactory perception. This model includes an AI-driven molecule generator (utilizing Graph and Generative Neural Networks), quantification and prediction of odor intensity, and refinery of optimal solvent and molecule combinations for desired fragrances. Additionally, a thermodynamic-based model establishes a link between olfactory perception and liquid-phase concentrations. The methodology employs Transfer Learning and selects the most suitable molecules based on vapor pressure and fragrance notes. Ultimately, a mathematical optimization problem is formulated to minimize discrepancies between new and target olfactory experiences. The methodology is validated by reproducing two distinct olfactory experiences using available experimental data.
Chemical Physics,Machine Learning
What problem does this paper attempt to address?
The paper attempts to address the problem of replicating and optimizing the olfactory experience of perfumes through machine learning methods. Specifically, the main objectives of the paper include: 1. **Establishing a link between molecular structure and human olfactory perception**: By developing a hybrid model that connects the molecular structure of perfumes with their impact on human olfaction. 2. **Generating and optimizing molecules**: Utilizing Graph Neural Networks and Generative Neural Networks to generate new molecules and predict their scent intensity. 3. **Selecting the best molecular combinations**: Based on vapor pressure and aroma characteristics, selecting the most suitable solvent and molecular combinations to achieve the desired fragrance. 4. **Minimizing the difference between the new fragrance and the target fragrance**: Using mathematical optimization methods to ensure that the newly generated perfume closely approximates the target scent. 5. **Reducing the time and cost of traditional trial-and-error methods**: By employing the aforementioned methods, reducing the time-consuming and expensive trial-and-error steps in the perfume development process. Overall, this research aims to propose a new framework that efficiently replicates and optimizes specific olfactory experiences by combining artificial intelligence and phenomenological knowledge. This not only aids innovation in the perfume industry but also promotes the preservation and study of olfactory heritage.