Odor prediction of whiskies based on their molecular composition

Abstract Aroma compositions are usually complex mixtures of odor-active compounds exhibiting diverse molecular structures. Due to chemical interactions of these compounds in the olfactory system, assessing or even predicting the olfactory quality of such mixtures is a difficult task, not only for st...

Full description

Saved in:
Bibliographic Details
Main Authors: Satnam Singh, Doris Schicker, Helen Haug, Tilman Sauerwald, Andreas T. Grasskamp
Format: Article
Language:English
Published: Nature Portfolio 2024-12-01
Series:Communications Chemistry
Online Access:https://doi.org/10.1038/s42004-024-01373-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850102194824544256
author Satnam Singh
Doris Schicker
Helen Haug
Tilman Sauerwald
Andreas T. Grasskamp
author_facet Satnam Singh
Doris Schicker
Helen Haug
Tilman Sauerwald
Andreas T. Grasskamp
author_sort Satnam Singh
collection DOAJ
description Abstract Aroma compositions are usually complex mixtures of odor-active compounds exhibiting diverse molecular structures. Due to chemical interactions of these compounds in the olfactory system, assessing or even predicting the olfactory quality of such mixtures is a difficult task, not only for statistical models, but even for trained assessors. Here, we combine fast automated analytical assessment tools with human sensory data of 11 experienced panelists and machine learning algorithms. Using 16 previously analyzed whisky samples (American or Scotch origin), we apply the linear classifier OWSum to distinguish the samples based on their detected molecules and to gain insights into the key molecular structure characteristics and odor descriptors for sample type. Moreover, we use OWSum and a Convolutional Neural Network (CNN) architecture to classify the five most relevant odor attributes of each sample and predict their sensory scores with promising accuracies (up to F1: 0.71, MCC: 0.68, ROCAUC: 0.78). The predictions outperform the inter-panelist agreement and thus demonstrate previously impossible data-driven sensory assessment in mixtures.
format Article
id doaj-art-86b476f478a0486184a7d5aa9a6d1121
institution DOAJ
issn 2399-3669
language English
publishDate 2024-12-01
publisher Nature Portfolio
record_format Article
series Communications Chemistry
spelling doaj-art-86b476f478a0486184a7d5aa9a6d11212025-08-20T02:39:48ZengNature PortfolioCommunications Chemistry2399-36692024-12-01711910.1038/s42004-024-01373-2Odor prediction of whiskies based on their molecular compositionSatnam Singh0Doris Schicker1Helen Haug2Tilman Sauerwald3Andreas T. Grasskamp4Department of Sensory Analytics and Technologies, Fraunhofer Institute for Process Engineering and Packaging IVVDepartment of Sensory Analytics and Technologies, Fraunhofer Institute for Process Engineering and Packaging IVVDepartment of Sensory Analytics and Technologies, Fraunhofer Institute for Process Engineering and Packaging IVVDepartment of Sensory Analytics and Technologies, Fraunhofer Institute for Process Engineering and Packaging IVVDepartment of Sensory Analytics and Technologies, Fraunhofer Institute for Process Engineering and Packaging IVVAbstract Aroma compositions are usually complex mixtures of odor-active compounds exhibiting diverse molecular structures. Due to chemical interactions of these compounds in the olfactory system, assessing or even predicting the olfactory quality of such mixtures is a difficult task, not only for statistical models, but even for trained assessors. Here, we combine fast automated analytical assessment tools with human sensory data of 11 experienced panelists and machine learning algorithms. Using 16 previously analyzed whisky samples (American or Scotch origin), we apply the linear classifier OWSum to distinguish the samples based on their detected molecules and to gain insights into the key molecular structure characteristics and odor descriptors for sample type. Moreover, we use OWSum and a Convolutional Neural Network (CNN) architecture to classify the five most relevant odor attributes of each sample and predict their sensory scores with promising accuracies (up to F1: 0.71, MCC: 0.68, ROCAUC: 0.78). The predictions outperform the inter-panelist agreement and thus demonstrate previously impossible data-driven sensory assessment in mixtures.https://doi.org/10.1038/s42004-024-01373-2
spellingShingle Satnam Singh
Doris Schicker
Helen Haug
Tilman Sauerwald
Andreas T. Grasskamp
Odor prediction of whiskies based on their molecular composition
Communications Chemistry
title Odor prediction of whiskies based on their molecular composition
title_full Odor prediction of whiskies based on their molecular composition
title_fullStr Odor prediction of whiskies based on their molecular composition
title_full_unstemmed Odor prediction of whiskies based on their molecular composition
title_short Odor prediction of whiskies based on their molecular composition
title_sort odor prediction of whiskies based on their molecular composition
url https://doi.org/10.1038/s42004-024-01373-2
work_keys_str_mv AT satnamsingh odorpredictionofwhiskiesbasedontheirmolecularcomposition
AT dorisschicker odorpredictionofwhiskiesbasedontheirmolecularcomposition
AT helenhaug odorpredictionofwhiskiesbasedontheirmolecularcomposition
AT tilmansauerwald odorpredictionofwhiskiesbasedontheirmolecularcomposition
AT andreastgrasskamp odorpredictionofwhiskiesbasedontheirmolecularcomposition