Predicting odor profile of food from its chemical composition: Towards an approach based on artificial intelligence and flavorists expertise.

Odor is central to food quality. Still, a major challenge is to understand how the odorants present in a given food contribute to its specific odor profile, and how to predict this olfactory outcome from the chemical composition. In this proof-of-concept study, we seek to develop an integrative model that combines expert knowledge, fuzzy logic, and machine learning to predict the quantitative odor description of complex mixtures of odorants. The model output is the intensity of relevant odor sensory attributes calculated on the basis of the content in odor-active comounds. The core of the model is the mathematically formalized knowledge of four senior flavorists, which provided a set of optimized rules describing the sensory-relevant combinations of odor qualities the experts have in mind to elaborate the target odor sensory attributes. The model first queries analytical and sensory databases in order to standardize, homogenize, and quantitatively code the odor descriptors of the odorants. Then the standardized odor descriptors are translated into a limited number of odor qualities used by the experts thanks to an ontology. A third step consists of aggregating all the information in terms of odor qualities across all the odorants found in a given product. The final step is a set of knowledge-based fuzzy membership functions representing the flavorist expertise and ensuring the prediction of the intensity of the target odor sensory descriptors on the basis of the products' aggregated odor qualities; several methods of optimization of the fuzzy membership functions have been tested. Finally, the model was applied to predict the odor profile of 16 red wines from two grape varieties for which the content in odorants was available. The results showed that the model can predict the perceptual outcome of food odor with a certain level of accuracy, and may also provide insights into combinations of odorants not mentioned by the experts.

Representation of rewards differing in their hedonic valence in the caudate nucleus correlates with the performance in a problem-solving task in dogs (Canis familiaris).

We have investigated dogs' (Canis familiaris) abilities in associating different sounds with appetitive stimuli of different incentive values. The association's establishment was first tested on family dogs (n = 20) in a problem-solving behavioural paradigm (experiment 1), then in a problem-solving behavioural paradigm as well as an fMRI study on specially trained family dogs (n = 20) (experiment 2). The aim was to show behavioural and parallel neural effects of the association formed between the two sounds and two different associated appetitive stimuli. The latency of solving the problem was considered an indicator of the motivational state. In our first experiment, where only behaviour was studied, we found that dogs were quicker in solving a problem upon hearing the sound associated with food higher in reward value, suggesting that they have successfully associated the sounds with the corresponding food value. In our second experiment, this behaviour difference was not significant. In the fMRI study, the cerebral response to the two sounds was compared both before and after the associative training. Two bilateral regions of interest were explored: the caudate nucleus and the amygdala. After the associative training, the response in the caudate nucleus was higher to the sound related to a higher reward value food than to the sound related to a lower reward value food, which difference was not present before the associative training. We found an increase in the amygdala response to both sounds after the training. In a whole-brain representational similarity analysis, we found that cerebral patterns in the caudate nucleus to the two sounds were different only after the training. Moreover, we found a positive correlation between the dissimilarity index in the caudate nucleus for activation responses to the two sounds and the difference in latencies (i.e. high reward value associated sound condition latency-low reward value associated sound condition latency) to solve the behavioural task: the bigger the difference between the conditions in latency to solve the task, the greater the difference in the neural representation of the two sounds was. In summary, family dogs' brain activation patterns reflected their expectations based on what they learned about the relationship between two sounds and their associated appetitive stimuli.

OOPS, the Ontology for Odor Perceptual Space: From Molecular Composition to Sensory Attributes of Odor Objects.

When creating a flavor to elicit a specific odor object characterized by odor sensory attributes (OSA), expert perfumers or flavorists use mental combinations of odor qualities (OQ) such as Fruity, Green, and Smoky. However, OSA and OQ are not directly related to the molecular composition in terms of odorants that constitute the chemical stimuli supporting odor object perception because of the complex non-linear integration of odor mixtures within the olfactory system. Indeed, single odorants are described with odor descriptors (OD), which can be found in various databases. Although classifications and aroma wheels studied the relationships between OD and OQ, the results were highly dependent on the studied products. Nevertheless, ontologies have proven to be very useful in sharing concepts across applications in a generic way and to allow experts' knowledge integration, implying non-linear cognitive processes. In this paper, we constructed the Ontology for Odor Perceptual Space (OOPS) to merge OD into a set of OQ best characterizing the odor, further translated into a set of OSA thanks to expert knowledge integration. Results showed that OOPS can help bridge molecular composition to odor perception and description, as demonstrated in the case of wines.

Comprehensive sensory and chemical data on the flavor of 16 red wines from two varieties: Sensory descriptive analysis, HS-SPME-GC-MS volatile compounds quantitative analysis, and odor-active compounds identification by HS-SPME-GC-MS-O.

This paper describes data collected on 2 sets of 8 French red wines from two grape varieties: Pinot Noir (PN) and Cabernet Franc (CF). It provides, for the 16 wines, (i) sensory descriptive data obtained with a trained panel, (ii) volatile organic compounds (VOC) quantification data obtained by Headspace Solid Phase Micro-Extraction - Gas Chromatography - Mass Spectrometry (HS-SPME-GC-MS) and (iii) odor-active compounds identification by Headspace Solid Phase Micro-Extraction - Gas Chromatography - Mass Spectrometry - Olfactometry (HS-SPME-GC-MS-O). The raw data are hosted on an open-access research data repository [1].

Quantitative Analysis of MicroRNAs in Vaccinia virus Infection Reveals Diversity in Their Susceptibility to Modification and Suppression.

Vaccinia virus (VACV) is a large cytoplasmic DNA virus that causes dramatic alterations to many cellular pathways including microRNA biogenesis. The virus encodes a poly(A) polymerase which was previously shown to add poly(A) tails to the 3' end of cellular miRNAs, resulting in their degradation by 24 hours post infection (hpi). Here we used small RNA sequencing to quantify the impact of VACV infection on cellular miRNAs in human cells at both early (6 h) and late (24 h) times post infection. A detailed quantitative analysis of individual miRNAs revealed marked diversity in the extent of their modification and relative change in abundance during infection. Some miRNAs became highly modified (e.g. miR-29a-3p, miR-27b-3p) whereas others appeared resistant (e.g. miR-16-5p). Furthermore, miRNAs that were highly tailed at 6 hpi were not necessarily among the most reduced at 24 hpi. These results suggest that intrinsic features of human cellular miRNAs cause them to be differentially polyadenylated and altered in abundance during VACV infection. We also demonstrate that intermediate and late VACV gene expression are required for optimal repression of some miRNAs including miR-27-3p. Overall this work reveals complex and varied consequences of VACV infection on host miRNAs and identifies miRNAs which are largely resistant to VACV-induced polyadenylation and are therefore present at functional levels during the initial stages of infection and replication.