Multivariate Statistical Analysis and Odor–Taste Network To Reveal Odor–Taste Associations

• Published in: Journal of agricultural and food chemistry Vol. 68; no. 38; pp. 10318 - 10328
• Main Authors: Guichard, Elisabeth, Barba, Carmen, Thomas-Danguin, Thierry, Tromelin, Anne
• Format: Journal Article
• Language: English
• Published: United States American Chemical Society 23.09.2020
• Subjects: Chemical Sciences; Flavoring Agents - analysis; Food and Nutrition; Humans; Life Sciences; Multivariate Analysis; Odorants - analysis; or physical chemistry; Smell; Taste; Theoretical and; odor−taste association; partial least square analysis; saltiness; bitterness; sweetness; sourness; multidimensional scaling; multivariate analysis; odor descriptors; odor-taste association
• ISSN: 0021-8561; 1520-5118
• DOI: 10.1021/acs.jafc.9b05462

Odor–taste association has been successfully applied to enhance taste perception in foods with low sugar or low salt content. Nevertheless, selecting odor descriptors with a given associated taste remains a challenge. In the aim to look for odors able to enhance some specific taste, we tested different multivariate analyses to find links between taste descriptors and odor descriptors, starting from a set of data previously obtained using gas chromatography/olfactometry-associated taste: 68 odorant zones described with 41 odor descriptors and 4 taste-associated descriptors (sweetness, saltiness, bitterness, and sourness). A partial least square analysis allowed for identification of odors associated with a specific taste. For instance, odors described as either fruity, sweet, strawberry, candy, floral, or orange are associated with sweetness, while odors described as either toasted, potato, sulfur, or mushroom are associated with saltiness. A network representation allowed for visualization of the links between odor and taste descriptors. As an example, a positive association was found between butter odor and both saltiness and sweetness. Our approach provided a visualization tool of the links between odor and taste description and could be used to select odor-active molecules with a potential taste enhancement effect based on their odor descriptors.