Predicting odor perceptual similarity from odor structure

Navigating the Fragrance Space Using Graph Generative Models and Predicting Odors

We explore a suite of generative modeling 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 a 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 the broader adoption of our research across applications in fragrance discovery and olfactory research.

Deep Learning for Odor Prediction on Aroma-Chemical Blends

The application of deep-learning techniques to aroma chemicals has resulted in models that surpass those of human experts in predicting olfactory qualities. However, public research in this field has been limited to predicting the qualities of individual molecules, whereas in industry, perfumers and food scientists are often more concerned with blends of multiple molecules. In this paper, we apply both established and novel approaches to a data set we compiled, which consists of labeled pairs of molecules. We present graph neural network models that accurately predict the olfactory qualities emerging from blends of aroma chemicals along with an analysis of how variations in model architecture can significantly impact predictive performance.

Predicting human olfactory perception by odorant structure and receptor activation profile

Humans possess a remarkable ability to discriminate a wide range of odors with high precision. This process begins with olfactory receptors (ORs) detecting and responding to the molecular structures of odorants. Recent studies have aimed to associate the activity of a single OR to an odor descriptor or predict odor descriptors using 2D molecular representation. However, predicting a limited number of odor descriptors is insufficient to fully understand the widespread and elaborate olfactory perception process. Therefore, we conducted structure-activity relationship analyses for ORs of eugenol, vanillin, and structurally similar compounds, investigating the correlation between molecular structures, OR activity profiles, and perceptual odor similarity. Our results indicated that these structurally similar compounds primarily activated 6 ORs, and the activity profiles of these ORs correlated with their perception. This enabled the development of a prediction model for the eugenol-similarity score from OR activity profiles (coefficient of determination, R2 = 0.687). Furthermore, the molecular structures of odorants were represented as 3D shapes and pharmacophore fingerprints, considering the 3D structural similarities between various odorants with multiple conformations. These 3D shape and pharmacophore fingerprints could also predict the perceptual odor similarity (R2 = 0.514). Finally, we identified key molecular structural features that contributed to predicting sensory similarities between compounds structurally similar to eugenol and vanillin. Our models, which predict odor from OR activity profiles and similarities in the 3D structure of odorants, may aid in understanding olfactory perception by compressing the information from a vast number of odorants into the activity profiles of 400 ORs.

RETRACTED: The role of black soldier fly (BSF) in eliminating the putrid odor of organic waste and its product application - A comprehensive review

This article has been retracted: please see Elsevier Policy on Article Withdrawal (https://www.elsevier.com/about/policies/article-withdrawal). This article has been retracted at the request of Editor. Post-publication, the editor discovered suspicious changes in authorship between the original submission and the revised version of this paper. In summary Ali Ahmad, who had not been part of the authorship of the original submission, was added as the new and only Corresponding Author. Shahida Anusha Siddiqui remained as part of the authorship, but no longer as sole Corresponding Author. These changes to the revised paper without explanation and without exceptional approval by the journal editor are contrary to the journal policy on changes to authorship. Both authors were contacted for an explanation. Shahida Anusha Siddiqui confirmed that sole corresponding authorship should be bestowed upon Ali Ahmad, but otherwise failed to provide a satisfactory explanation for these changes. Ali Ahmad did not respond. Overall, the editor feels that the findings of the manuscript cannot be relied upon and that the article needs to be retracted. The journal apologises for not having identified the problematic authorship changes during the review process and for any resulting inconvenience.

Pyrfume: A window to the world's olfactory data

Advances in theoretical understanding are frequently unlocked by access to large, diverse experimental datasets. Our understanding of olfactory neuroscience and psychophysics remain years behind the other senses, in part because rich datasets linking olfactory stimuli with their corresponding percepts, behaviors, and neural pathways remain scarce. Here we present a concerted effort to unlock and unify dozens of stimulus-linked olfactory datasets across species and modalities under a unified framework called Pyrfume. We present examples of how researchers might use Pyrfume to conduct novel analyses uncovering new principles, introduce trainees to the field, or construct benchmarks for machine olfaction.

Decomposition of an odorant in olfactory perception and neural representation

Molecules-the elementary units of substances-are commonly considered the units of processing in olfactory perception, giving rise to undifferentiated odour objects invariant to environmental variations. By selectively perturbing the processing of chemical substructures with adaptation ('the psychologist's microelectrode') in a series of psychophysical and neuroimaging experiments (458 participants), we show that two perceptually distinct odorants sharing part of their structural features become significantly less discernible following adaptation to a third odorant containing their non-shared structural features, in manners independent of olfactory intensity, valence, quality or general olfactory adaptation. The effect is accompanied by reorganizations of ensemble activity patterns in the posterior piriform cortex that parallel subjective odour quality changes, in addition to substructure-based neural adaptations in the anterior piriform cortex and amygdala. Central representations of odour quality and the perceptual outcome thus embed submolecular structural information and are malleable by recent olfactory encounters.

Nature and human well-being: The olfactory pathway

The world is undergoing massive atmospheric and ecological change, driving unprecedented challenges to human well-being. Olfaction is a key sensory system through which these impacts occur. The sense of smell influences quality of and satisfaction with life, emotion, emotion regulation, cognitive function, social interactions, dietary choices, stress, and depressive symptoms. Exposures via the olfactory pathway can also lead to (anti-)inflammatory outcomes. Increased understanding is needed regarding the ways in which odorants generated by nature (i.e., natural olfactory environments) affect human well-being. With perspectives from a range of health, social, and natural sciences, we provide an overview of this unique sensory system, four consensus statements regarding olfaction and the environment, and a conceptual framework that integrates the olfactory pathway into an understanding of the effects of natural environments on human well-being. We then discuss how this framework can contribute to better accounting of the impacts of policy and land-use decision-making on natural olfactory environments and, in turn, on planetary health.

What Does the Human Olfactory System Do, and How Does It Do It?

Historically, the human sense of smell has been regarded as the odd stepchild of the senses, especially compared to the sensory bravado of seeing, touching, and hearing. The idea that the human olfaction has little to contribute to our experience of the world is commonplace, though with the emergence of COVID-19 there has rather been a sea change in this understanding. An ever increasing body of work has convincingly highlighted the keen capabilities of the human nose and the sophistication of the human olfactory system. Here, we provide a concise overview of the neuroscience of human olfaction spanning the last 10-15 years, with focus on the peripheral and central mechanisms that underlie how odor information is processed, packaged, parceled, predicted, and perturbed to serve odor-guided behaviors. We conclude by offering some guideposts for harnessing the next decade of olfactory research in all its shapes and forms.

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.

Perceptual metrics for odorants: Learning from non-expert similarity feedback using machine learning

Defining perceptual similarity metrics for odorant comparisons is crucial to understanding the mechanism of olfactory perception. Current methods in olfaction rely on molecular physicochemical features or discrete verbal descriptors (floral, burnt, etc.) to approximate perceptual (dis)similarity between odorants. However, structural or verbal descriptors alone are limited in modeling complex nuances of odor perception. While structural features inadequately characterize odor perception, language-based discrete descriptors lack the granularity needed to model a continuous perception space. We introduce data-driven approaches to perceptual metrics learning (PMeL) based on two key insights: a) by combining physicochemical features with the user's perceptual feedback, we can leverage both structural and perceptual attributes of odors to define dissimilarity, and b) instead of discrete labels, user's perceptual feedback can be gathered as relative similarity comparisons, such as "Does molecule-A smell more like molecule-B, or molecule-C?" These triplet comparisons are easier even for non-experts users and offer a more effective representation of the continuous perception space. Experimental results on several defined tasks show the effectiveness of our approach in evaluating perceptual dissimilarity between odorants. Finally, we investigate how closely our model, trained on non-expert feedback, aligns with the expert's similarity judgments. Our effort aims to reduce reliance on expert annotations.

iORbase: A database for the prediction of the structures and functions of insect olfactory receptors

Insect olfactory receptors (iORs) with atypical 7-transmembrane domains, unlike Chordata olfactory receptors, are not in the GPCR protein family. iORs selectively bind to volatile ligands in the environment and affect essential insect behaviors. In this study, we constructed a new platform (iORbase, https://www.iorbase.com) for the structural and functional analysis of iORs based on a combined algorithm for gene annotation and protein structure prediction. Moreover, it provides the option to calculate the binding affinities and binding residues between iORs and pheromone molecules by virtual screening of docking. Furthermore, iORbase supports the automatic structural and functional prediction of user-submitted iORs or pheromones. iORbase contains the well-analyzed results of approximately 6 000 iORs and their 3D protein structures identified from 59 insect species and 2 077 insect pheromones from the literature, as well as approximately 12 million pairs of simulated interactions between functional iORs and pheromones. We also built 4 online modules, iORPDB, iInteraction, iModelTM, and iOdorTool to easily retrieve and visualize the 3D structures and interactions. iORbase can help greatly improve the experimental efficiency and success rate, identify new insecticide targets, or develop electronic nose technology. This study will shed light on the olfactory recognition mechanism and evolutionary characteristics from the perspectives of omics and macroevolution.

A principal odor map unifies diverse tasks in olfactory perception

Mapping molecular structure to odor perception is a key challenge in olfaction. We used graph neural networks to generate a principal odor map (POM) that preserves perceptual relationships and enables odor quality prediction for previously uncharacterized odorants. The model was as reliable as a human in describing odor quality: On a prospective validation set of 400 out-of-sample odorants, the model-generated odor profile more closely matched the trained panel mean than did the median panelist. By applying simple, interpretable, theoretically rooted transformations, the POM outperformed chemoinformatic models on several other odor prediction tasks, indicating that the POM successfully encoded a generalized map of structure-odor relationships. This approach broadly enables odor prediction and paves the way toward digitizing odors.

High-precision mapping reveals the structure of odor coding in the human brain

Odor perception is inherently subjective. Previous work has shown that odorous molecules evoke distributed activity patterns in olfactory cortices, but how these patterns map on to subjective odor percepts remains unclear. In the present study, we collected neuroimaging responses to 160 odors from 3 individual subjects (18 h per subject) to probe the neural coding scheme underlying idiosyncratic odor perception. We found that activity in the orbitofrontal cortex (OFC) represents the fine-grained perceptual identity of odors over and above coarsely defined percepts, whereas this difference is less pronounced in the piriform cortex (PirC) and amygdala. Furthermore, the implementation of perceptual encoding models enabled us to predict olfactory functional magnetic resonance imaging responses to new odors, revealing that the dimensionality of the encoded perceptual spaces increases from the PirC to the OFC. Whereas encoding of lower-order dimensions generalizes across subjects, encoding of higher-order dimensions is idiosyncratic. These results provide new insights into cortical mechanisms of odor coding and suggest that subjective olfactory percepts reside in the OFC.

Physicochemical features partially explain olfactory crossmodal correspondences

During the olfactory perception process, our olfactory receptors are thought to recognize specific chemical features. These features may contribute towards explaining our crossmodal perception. The physicochemical features of odors can be extracted using an array of gas sensors, also known as an electronic nose. The present study investigates the role that the physicochemical features of olfactory stimuli play in explaining the nature and origin of olfactory crossmodal correspondences, which is a consistently overlooked aspect of prior work. Here, we answer the question of whether the physicochemical features of odors contribute towards explaining olfactory crossmodal correspondences and by how much. We found a similarity of 49% between the perceptual and the physicochemical spaces of our odors. All of our explored crossmodal correspondences namely, the angularity of shapes, smoothness of textures, perceived pleasantness, pitch, and colors have significant predictors for various physicochemical features, including aspects of intensity and odor quality. While it is generally recognized that olfactory perception is strongly shaped by context, experience, and learning, our findings show that a link, albeit small (6-23%), exists between olfactory crossmodal correspondences and their underlying physicochemical features.

OWSum: algorithmic odor prediction and insight into structure-odor relationships

We derived and implemented a linear classification algorithm for the prediction of a molecule's odor, called Olfactory Weighted Sum (OWSum). Our approach relies solely on structural patterns of the molecules as features for algorithmic treatment and uses conditional probabilities combined with tf-idf values. In addition to the prediction of molecular odor, OWSum provides insights into properties of the dataset and allows to understand how algorithmic classifications are reached by quantitatively assigning structural patterns to odors. This provides chemists with an intuitive understanding of underlying interactions. To deal with ambiguities of the natural language used to describe odor, we introduced descriptor overlap as a metric for the quantification of semantic overlap between descriptors. Thus, grouping of descriptors and derivation of higher-level descriptors becomes possible. Our approach poses a large leap forward in our capabilities to understand and predict molecular features.

Olfactory abstraction: a communicative and metacognitive account

The usual puzzle raised about olfaction is that of a deficit of abstraction: smells, by contrast notably with colours, do not easily lend themselves to abstract categories and labels. Some studies have argued that the puzzle is culturally restricted and that abstraction is more common outside urban Western societies. Here, I argue that the puzzle is misconstrued and should be reversed: given that odours are constantly changing and that their commonalities are difficult for humans to identify, what is surprising is not that abstract terms are rare, but that they should be used at all for olfaction. Given the nature of the olfactory environment and our cognitive equipment, concrete labels referring to sources seem most adaptive. To explain the use and presence of abstract terms, we need to examine their social and communicative benefits. Here these benefits are spelt out as securing a higher agreement among individuals varying in their olfactory experiences as well as the labels they use, as well as feeling a heightened sense of confidence in one's naming capacities. This article is part of the theme issue 'Concepts in interaction: social engagement and inner experiences'.

Expansive linguistic representations to predict interpretable odor mixture discriminability

Language is often thought as being poorly adapted to precisely describe or quantify smell and olfactory attributes. In this work, we show that semantic descriptors of odors can be implemented in a model to successfully predict odor mixture discriminability, an olfactory attribute. We achieved this by taking advantage of the structure-to-percept model we previously developed for monomolecular odorants, using chemical descriptors to predict pleasantness, intensity and 19 semantic descriptors such as "fish," "cold," "burnt," "garlic," "grass," and "sweet" for odor mixtures, followed by a metric learning to obtain odor mixture discriminability. Through this expansion of the representation of olfactory mixtures, our Semantic model outperforms state of the art methods by taking advantage of the intermediary semantic representations learned from human perception data to enhance and generalize the odor discriminability/similarity predictions. As 10 of the semantic descriptors were selected to predict discriminability/similarity, our approach meets the need of rapidly obtaining interpretable attributes of odor mixtures as illustrated by the difficulty of finding olfactory metamers. More fundamentally, it also shows that language can be used to establish a metric of discriminability in the everyday olfactory space.

Hyperbolic odorant mixtures as a basis for more efficient signaling between flowering plants and bees

Animals use odors in many natural contexts, for example, for finding mates or food, or signaling danger. Most analyses of natural odors search for either the most meaningful components of a natural odor mixture, or they use linear metrics to analyze the mixture compositions. However, we have recently shown that the physical space for complex mixtures is ‘hyperbolic’, meaning that there are certain combinations of variables that have a disproportionately large impact on perception and that these variables have specific interpretations in terms of metabolic processes taking place inside the flower and fruit that produce the odors. Here we show that the statistics of odorants and odorant mixtures produced by inflorescences (Brassica rapa) are also better described with a hyperbolic rather than a linear metric, and that combinations of odorants in the hyperbolic space are better predictors of the nectar and pollen resources sought by bee pollinators than the standard Euclidian combinations. We also show that honey bee and bumble bee antennae can detect most components of the B. rapa odor space that we tested, and the strength of responses correlates with positions of odorants in the hyperbolic space. In sum, a hyperbolic representation can be used to guide investigation of how information is represented at different levels of processing in the CNS.

A behavioral paradigm for measuring perceptual distances in mice

Perceptual similarities between a specific stimulus and other stimuli of the same modality provide valuable information about the structure and geometry of sensory spaces. While typically assessed in human behavioral experiments, perceptual similarities-or distances-are rarely measured in other species. However, understanding the neural computations responsible for sensory representations requires the monitoring and often manipulation of neural activity, which is more readily achieved in non-human experimental models. Here, we develop a behavioral paradigm that enables the quantification of perceptual similarity between sensory stimuli using mouse olfaction as a model system.

Finding the Brain in the Nose

Olfaction is fundamentally distinct from other sensory modalities. Natural odor stimuli are complex mixtures of volatile chemicals that interact in the nose with a receptor array that, in rodents, is built from more than 1,000 unique receptors. These interactions dictate a peripheral olfactory code, which in the brain is transformed and reformatted as it is broadcast across a set of highly interconnected olfactory regions. Here we discuss the problems of characterizing peripheral population codes for olfactory stimuli, of inferring the specific functions of different higher olfactory areas given their extensive recurrence, and of ultimately understanding how odor representations are linked to perception and action. We argue that, despite the differences between olfaction and other sensory modalities, addressing these specific questions will reveal general principles underlying brain function.

A physicochemical model of odor sampling

We present a general physicochemical sampling model for olfaction, based on established pharmacological laws, in which arbitrary combinations of odorant ligands and receptors can be generated and their individual and collective effects on odor representations and olfactory performance measured. Individual odor ligands exhibit receptor-specific affinities and efficacies; that is, they may bind strongly or weakly to a given receptor, and can act as strong agonists, weak agonists, partial agonists, or antagonists. Ligands interacting with common receptors compete with one another for dwell time; these competitive interactions appropriately simulate the degeneracy that fundamentally defines the capacities and limitations of odorant sampling. The outcome of these competing ligand-receptor interactions yields a pattern of receptor activation levels, thereafter mapped to glomerular presynaptic activation levels based on the convergence of sensory neuron axons. The metric of greatest interest is the mean discrimination sensitivity, a measure of how effectively the olfactory system at this level is able to recognize a small change in the physicochemical quality of a stimulus. This model presents several significant outcomes, both expected and surprising. First, adding additional receptors reliably improves the system's discrimination sensitivity. Second, in contrast, adding additional ligands to an odorscene initially can improve discrimination sensitivity, but eventually will reduce it as the number of ligands increases. Third, the presence of antagonistic ligand-receptor interactions produced clear benefits for sensory system performance, generating higher absolute discrimination sensitivities and increasing the numbers of competing ligands that could be present before discrimination sensitivity began to be impaired. Finally, the model correctly reflects and explains the modest reduction in odor discrimination sensitivity exhibited by transgenic mice in which the specificity of glomerular targeting by primary olfactory neurons is partially disrupted.

A dataset on odor intensity and odor pleasantness of 222 binary mixtures of 72 key food odorants rated by a sensory panel of 30 trained assessors

This paper describes data collected on a set of 222 binary mixtures, based on a set of 72 odorants chiefly found in food, rated by 30 selected and trained assessors for odor intensity and pleasantness. The data included odor intensity (IAB) and pleasantness (PAB) of the mixtures, the intensity (IA, IB) and the pleasantness (PA, PB) of the two components. Moreover, the intensity (IAmix, IBmix) of the two components’ odor perceived within the mixture are included. The quality of the dataset was evaluated by checking subjects’ performance and by testing repeatability using the 24 duplicated trials for each attribute. This set of experimental data would be especially valuable to investigate theories of odor mixture perception in human and to test new models to predict odor perception of odor mixtures.

Olfactory Perception in Relation to the Physicochemical Odor Space

A growing body of research aims at solving what is often referred to as the stimulus-percept problem in olfactory perception. Although computational efforts have made it possible to predict perceptual impressions from the physicochemical space of odors, studies with large psychophysical datasets from non-experts remain scarce. Following previous approaches, we developed a physicochemical odor space using 4094 molecular descriptors of 1389 odor molecules. For 20 of these odors, we examined associations with perceived pleasantness, intensity, odor quality and detection threshold, obtained from a dataset of 2000 naïve participants. Our results show significant differences in perceptual ratings, and we were able to replicate previous findings on the association between perceptual ratings and the first dimensions of the physicochemical odor space. However, the present analyses also revealed striking interindividual variations in perceived pleasantness and intensity. Additionally, interactions between pleasantness, intensity, and olfactory and trigeminal qualitative dimensions were found. To conclude, our results support previous findings on the relation between structure and perception on the group level in our sample of non-expert raters. In the challenging task to relate olfactory stimulus and percept, the physicochemical odor space can serve as a reliable and helpful tool to structure the high-dimensional space of olfactory stimuli. Nevertheless, human olfactory perception in the individual is not an analytic process of molecule detection alone, but is part of a holistic integration of multisensory inputs, context and experience.

Parsing Sage and Rosemary in Time: The Machine Learning Race to Crack Olfactory Perception

Color and pitch perception are largely understandable from characteristics of physical stimuli: the wavelengths of light and sound waves, respectively. By contrast, understanding olfactory percepts from odorous stimuli (volatile molecules) is much more challenging. No intuitive set of molecular features is up to the task. Here in Chemical Senses, the Ray lab reports using a predictive modeling framework-first breaking molecular structure into thousands of features and then using this to train a predictive statistical model on a wide range of perceptual descriptors-to create a tool for predicting the odor character of hundreds of thousands of available but previously uncharacterized molecules (Kowalewski et al. 2021). This will allow future investigators to representatively sample the space of odorous molecules as well as identify previously unknown odorants with a target odor character. Here, I put this work into the context of other modeling efforts and highlight the urgent need for large new datasets and transparent benchmarks for the field to make and evaluate modeling breakthroughs, respectively.

A System-Wide Understanding of the Human Olfactory Percept Chemical Space

The fundamental units of olfactory perception are discrete 3D structures of volatile chemicals that each interact with specific subsets of a very large family of hundreds of odorant receptor proteins, in turn activating complex neural circuitry and posing a challenge to understand. We have applied computational approaches to analyze olfactory perceptual space from the perspective of odorant chemical features. We identify physicochemical features associated with ~150 different perceptual descriptors and develop machine-learning models. Validation of predictions shows a high success rate for test set chemicals within a study, as well as across studies more than 30 years apart in time. Due to the high success rates, we are able to map ~150 percepts onto a chemical space of nearly 0.5 million compounds, predicting numerous percept-structure combinations. The chemical structure-to-percept prediction provides a system-level view of human olfaction and opens the door for comprehensive computational discovery of fragrances and flavors.

Categorizing Smells: A Localist Approach

Humans are poorer at identifying smells and communicating about them, compared to other sensory domains. They also cannot easily organize odor sensations in a general conceptual space, where geometric distance could represent how similar or different all odors are. These two generalities are more or less accepted by psychologists, and they are often seen as connected: If there is no conceptual space for odors, then olfactory identification should indeed be poor. We propose here an important revision to this conclusion: We believe that the claim that there is no odor space is true only if by odor space, one means a conceptual space representing all possible odor sensations, in the paradigmatic sense used for instance for color. However, in a less paradigmatic sense, local conceptual spaces representing a given subset of odors do exist. Thus the absence of a global odor space does not warrant the conclusion that there is no olfactory conceptual map at all. Here we show how a localist account provides a new interpretation of experts and cross-cultural categorization studies: Rather than being exceptions to the poor olfactory identification and communication usually seen elsewhere, experts and cross-cultural categorization are here taken to corroborate the existence of local conceptual spaces.

More Data, Please: Machine Learning to Advance the Multidisciplinary Science of Human Sociochemistry

Communication constitutes the core of human life. A large portion of our everyday social interactions is non-verbal. Of the sensory modalities we use for non-verbal communication, olfaction (i.e., the sense of smell) is often considered the most enigmatic medium. Outside of our awareness, smells provide information about our identity, emotions, gender, mate compatibility, illness, and potentially more. Yet, body odors are astonishingly complex, with their composition being influenced by various factors. Is there a chemical basis of olfactory communication? Can we identify molecules predictive of psychological states and traits? We propose that answering these questions requires integrating two disciplines: psychology and chemistry. This new field, coined sociochemistry, faces new challenges emerging from the sheer amount of factors causing variability in chemical composition of body odorants on the one hand (e.g., diet, hygiene, skin bacteria, hormones, genes), and variability in psychological states and traits on the other (e.g., genes, culture, hormones, internal state, context). In past research, the reality of these high-dimensional data has been reduced in an attempt to isolate unidimensional factors in small, homogenous samples under tightly controlled settings. Here, we propose big data approaches to establish novel links between chemical and psychological data on a large scale from heterogeneous samples in ecologically valid settings. This approach would increase our grip on the way chemical signals non-verbally and subconsciously affect our social lives across contexts.

Predicting brain organization with a computational model: 50-year perspective on lateral inhibition and oscillatory gating by dendrodendritic synapses

The first compartmental computer models of brain neurons using the Rall method predicted novel and unexpected dendrodendritic interactions between mitral and granule cells in the olfactory bulb. We review the models from a 50-year perspective on the work that has challenged, supported, and extended the original proposal that these interactions mediate both lateral inhibition and oscillatory activity, essential steps in the neural basis of olfactory processing and perception. We highlight strategies behind the neurophysiological experiments and the Rall methods that enhance the ability of detailed compartmental modeling to give counterintuitive predictions that lead to deeper insights into neural organization at the synaptic and circuit level. The application of these methods to mechanisms of neurogenesis and plasticity are exciting challenges for the future.

Manipulating synthetic optogenetic odors reveals the coding logic of olfactory perception

How does neural activity generate perception? Finding the combinations of spatial or temporal activity features (such as neuron identity or latency) that are consequential for perception remains challenging. We trained mice to recognize synthetic odors constructed from parametrically defined patterns of optogenetic activation, then measured perceptual changes during extensive and controlled perturbations across spatiotemporal dimensions. We modeled recognition as the matching of patterns to learned templates. The templates that best predicted recognition were sequences of spatially identified units, ordered by latencies relative to each other (with minimal effects of sniff). Within templates, individual units contributed additively, with larger contributions from earlier-activated units. Our synthetic approach reveals the fundamental logic of the olfactory code and provides a general framework for testing links between sensory activity and perception.

Compounds without borders: A mechanism for quantifying complex odors and responses to scent-pollution in bumblebees

Bumblebees are critical pollinators whose populations have been experiencing troubling declines over the past several decades. Successful foraging improves colony fitness, thus understanding how anthropogenic influences modulate foraging behavior may aid conservation efforts. Odor pollution can have negative impacts on bumble- and honey-bees foraging behavior. However, given the vast array of potential scent contaminants, individually testing pollutants is an ineffective approach. The ability to quantitatively measure how much scent-pollution of a floral-odor bumblebees can tolerate would represent a paradigm shift in odor-pollution studies. Current statistical methods for analyzing complex odors have poor predictive power because statistically-derived odor-spaces are rewritten when new odors are added. This study presents an alternative method of analyzing complex odor blends based on the encoding properties of insect olfactory systems. This “Compounds Without Borders” (CWB) method vectorizes odors in a multidimensional space representing relevant functional group and carbon characteristics of their component odorants. A single vector can be built for any scent, which allows the angular distance between any two odors to be calculated–including a learned odor and its polluted counterpart. Data presented here indicate that CWB-angles are capable of both describing and predicting bumblebee odor-discrimination behavior: odor pairs with angular distances in the 20–29° range appear to be generalized, while odor pairs over 30 degrees are differentiated. The neurophysiological properties underlying CWB-vectorization of odors are not unique to bumblebees; CWB-angle analysis of a small sample of published odor-data supports the idea that this method may have broader applications.

The giant pouched rat (Cricetomys ansorgei) olfactory receptor repertoire

For rodents, olfaction is essential for locating food, recognizing mates and competitors, avoiding predators, and navigating their environment. It is thought that rodents may have expanded olfactory receptor repertoires in order to specialize in olfactory behavior. Despite being the largest clade of mammals and depending on olfaction relatively little work has documented olfactory repertoires outside of conventional laboratory species. Here we report the olfactory receptor repertoire of the African giant pouched rat (Cricetomys ansorgei), a Muroid rodent distantly related to mice and rats. The African giant pouched rat is notable for its large cortex and olfactory bulbs relative to its body size compared to other sympatric rodents, which suggests anatomical elaboration of olfactory capabilities. We hypothesized that in addition to anatomical elaboration for olfaction, these pouched rats might also have an expanded olfactory receptor repertoire to enable their olfactory behavior. We examined the composition of the olfactory receptor repertoire to better understand how their sensory capabilities have evolved. We identified 1145 intact olfactory genes, and 260 additional pseudogenes within 301 subfamilies from the African giant pouched rat genome. This repertoire is similar to mice and rats in terms of size, pseudogene percentage and number of subfamilies. Analyses of olfactory receptor gene trees revealed that the pouched rat has 6 expansions in different subfamilies compared to mice, rats and squirrels. We identified 81 orthologous genes conserved among 4 rodent species and an additional 147 conserved genes within the Muroid rodents. The orthologous genes shared within Muroidea suggests that there may be a conserved Muroid-specific olfactory receptor repertoire. We also note that the description of this repertoire can serve as a complement to other studies of rodent olfaction, as the pouched rat is an outgroup within Muroidea. Thus, our data suggest that African giant pouched rats are capable of both natural and trained olfactory behaviors with a typical Muriod olfactory receptor repertoire.

SmellSpace: An Odor-Based Social Network as a Platform for Collecting Olfactory Perceptual Data

A common goal in olfaction research is modeling the link between odorant structure and odor perception. Such modeling efforts require large data sets on olfactory perception, yet only a few of these are publicly and freely available. Given that individual odor perception may be informative on personal makeup and interpersonal relationships, we hypothesized that people would gladly provide olfactory perceptual estimates in the context of an odor-based social network. We developed a web-based infrastructure for such a network we called SmellSpace and distributed 10 000 scratch-and-sniff registration booklets each containing a subset of 12 out of 35 microencapsulated odorants. Within ~100 days, we obtained data from ~1000 participants who rated the odorants along 13 verbal descriptors. To verify that these estimates are comparable to lab-collected estimates we tested 26 participants in a controlled lab setting using the same odorants and descriptors. We observed remarkably high overall group correlations between lab and SmellSpace data, implying that this method provides for credible group-representations of odorants. We further estimated the usability of the data by applying to it two previously published models that used odorant structure alone to predict either odorant pleasantness or pairwise odorant perceptual similarity. We observed statistically significant predictions in both cases, thus further implying that the current data may be helpful toward future efforts of modeling olfactory perception from structure. We conclude that an odor-based social network is a potentially useful instrument for collecting extensive data on olfactory perception and here post the complete raw data set from the first ~1000 participants.

The contribution of temporal coding to odor coding and odor perception in humans

Whether neurons encode information through their spike rates, their activity times or both is an ongoing debate in systems neuroscience. Here, we tested whether humans can discriminate between a pair of temporal odor mixtures (TOMs) composed of the same two components delivered in rapid succession in either one temporal order or its reverse. These TOMs presumably activate the same olfactory neurons but at different times and thus differ mainly in the time of neuron activation. We found that most participants could hardly discriminate between TOMs, although they easily discriminated between a TOM and one of its components. By contrast, participants succeeded in discriminating between the TOMs when they were notified of their successive nature in advance. We thus suggest that the time of glomerulus activation can be exploited to extract odor-related information, although it does not change the odor perception substantially, as should be expected from an odor code per se.

Odors: from chemical structures to gaseous plumes

We are immersed within an odorous sea of chemical currents that we parse into individual odors with complex structures. Odors have been posited as determined by the structural relation between the molecules that compose the chemical compounds and their interactions with the receptor site. But, naturally occurring smells are parsed from gaseous odor plumes. To give a comprehensive account of the nature of odors the chemosciences must account for these large distributed entities as well. We offer a focused review of what is known about the perception of odor plumes for olfactory navigation and tracking, which we then connect to what is known about the role odorants play as properties of the plume in determining odor identity with respect to odor quality. We end by motivating our central claim that more research needs to be conducted on the role that odorants play within the odor plume in determining odor identity.

Odor quality profile is partially influenced by verbal cues

Characterizing an odor quality is difficult for humans. Ever-increasing physiological and behavioral studies have characterized odor quality and demonstrated high performance of human odor categorization. However, there are no precise methods for measuring the multidimensional axis of an odor quality. Furthermore, it can be altered by individual experience, even when using existing measurement methods for the multidimensional axis of odor such as odor profiling. It is, therefore, necessary to characterize patterns of odor quality with odor profiling and observe alterations in odor profiles under the influence of subjective rating conditions such as verbal cues. Considering the high performance of human odor categorization, we hypothesized that odor may have specific odor quality that is scarcely altered by verbal cues. We assessed odor responses to isovaleric acid with and without verbal cues and compared the results in each stimulation condition. We found that verbal cues influenced the rating of odor quality descriptors. Verbal cues weakly influenced the odor quality descriptors of high-rated value (upper 25%) compared to odor quality descriptors of low-rated value (lower 75%) by the survey test. Even under different verbal cue conditions, the same odor was classified in the same class when using high-rated odor quality descriptors. Our study suggests that people extract essential odor quality descriptors that represent the odor itself in order to efficiently quantify odor quality.

An argument for hyperbolic geometry in neural circuits

This review connects several lines of research to argue that hyperbolic geometry should be broadly applicable to neural circuits as well as other biological circuits. The reason for this is that networks that conform to hyperbolic geometry are maximally responsive to external and internal perturbations. These networks also allow for efficient communication under conditions where nodes are added or removed. We will argue that one of the signatures of hyperbolic geometry is the celebrated Zipf's law (also sometimes known as the Pareto distribution) that states that the probability to observe a given pattern is inversely related to its rank. Zipf's law is observed in a variety of biological systems - from protein sequences, neural networks to economics. These observations provide further evidence for the ubiquity of networks with an underlying hyperbolic metric structure. Recent studies in neuroscience specifically point to the relevance of a three-dimensional hyperbolic space for neural signaling. The three-dimensional hyperbolic space may confer additional robustness compared to other dimensions. We illustrate how the use of hyperbolic coordinates revealed a novel topographic organization within the olfactory system. The use of such coordinates may facilitate representation of relevant signals elsewhere in the brain.

Chemical features mining provides new descriptive structure-odor relationships

An important goal in researching the biology of olfaction is to link the perception of smells to the chemistry of odorants. In other words, why do some odorants smell like fruits and others like flowers? While the so-called stimulus-percept issue was resolved in the field of color vision some time ago, the relationship between the chemistry and psycho-biology of odors remains unclear up to the present day. Although a series of investigations have demonstrated that this relationship exists, the descriptive and explicative aspects of the proposed models that are currently in use require greater sophistication. One reason for this is that the algorithms of current models do not consistently consider the possibility that multiple chemical rules can describe a single quality despite the fact that this is the case in reality, whereby two very different molecules can evoke a similar odor. Moreover, the available datasets are often large and heterogeneous, thus rendering the generation of multiple rules without any use of a computational approach overly complex. We considered these two issues in the present paper. First, we built a new database containing 1689 odorants characterized by physicochemical properties and olfactory qualities. Second, we developed a computational method based on a subgroup discovery algorithm that discriminated perceptual qualities of smells on the basis of physicochemical properties. Third, we ran a series of experiments on 74 distinct olfactory qualities and showed that the generation and validation of rules linking chemistry to odor perception was possible. Taken together, our findings provide significant new insights into the relationship between stimulus and percept in olfaction. In addition, by automatically extracting new knowledge linking chemistry of odorants and psychology of smells, our results provide a new computational framework of analysis enabling scientists in the field to test original hypotheses using descriptive or predictive modeling.

An important issue in olfaction sciences deals with the question of how a chemical information can be translated into percepts. This is known as the stimulus-percept problem. Here, we set out to better understand this issue by combining knowledge about the chemistry and cognition of smells with computational olfaction. We also assumed that not only one, but several physicochemical models may describe a given olfactory quality. To achieve this aim, a first challenge was to set up a database with ~1700 molecules characterized by chemical features and described by olfactory qualities (e.g. fruity, woody). A second challenge consisted in developing a computational model enabling the discrimination of olfactory qualities based on these chemical features. By meeting these 2 challenges, we provided for several olfactory qualities new chemical models describing why an odorant molecule smells fruity or woody (among others). For most qualities, multiple (rather than a single) chemical models were generated. These findings provide new elements of knowledge about the relationship between odorant chemistry and perception. They also make it possible to envisage concrete applications in the aroma and fragrance field where chemical characterization of smells is an important step in the design of new products.

Adaptation of olfactory receptor abundances for efficient coding

Olfactory receptor usage is highly heterogeneous, with some receptor types being orders of magnitude more abundant than others. We propose an explanation for this striking fact: the receptor distribution is tuned to maximally represent information about the olfactory environment in a regime of efficient coding that is sensitive to the global context of correlated sensor responses. This model predicts that in mammals, where olfactory sensory neurons are replaced regularly, receptor abundances should continuously adapt to odor statistics. Experimentally, increased exposure to odorants leads variously, but reproducibly, to increased, decreased, or unchanged abundances of different activated receptors. We demonstrate that this diversity of effects is required for efficient coding when sensors are broadly correlated, and provide an algorithm for predicting which olfactory receptors should increase or decrease in abundance following specific environmental changes. Finally, we give simple dynamical rules for neural birth and death processes that might underlie this adaptation.

Accurate prediction of personalized olfactory perception from large-scale chemoinformatic features

Background

The olfactory stimulus-percept problem has been studied for more than a century, yet it is still hard to precisely predict the odor given the large-scale chemoinformatic features of an odorant molecule. A major challenge is that the perceived qualities vary greatly among individuals due to different genetic and cultural backgrounds. Moreover, the combinatorial interactions between multiple odorant receptors and diverse molecules significantly complicate the olfaction prediction. Many attempts have been made to establish structure-odor relationships for intensity and pleasantness, but no models are available to predict the personalized multi-odor attributes of molecules. In this study, we describe our winning algorithm for predicting individual and population perceptual responses to various odorants in the DREAM Olfaction Prediction Challenge.

Results

We find that random forest model consisting of multiple decision trees is well suited to this prediction problem, given the large feature spaces and high variability of perceptual ratings among individuals. Integrating both population and individual perceptions into our model effectively reduces the influence of noise and outliers. By analyzing the importance of each chemical feature, we find that a small set of low- and nondegenerative features is sufficient for accurate prediction.

Conclusions

Our random forest model successfully predicts personalized odor attributes of structurally diverse molecules. This model together with the top discriminative features has the potential to extend our understanding of olfactory perception mechanisms and provide an alternative for rational odorant design.

Hyperbolic geometry of the olfactory space

In the natural environment, the sense of smell, or olfaction, serves to detect toxins and judge nutritional content by taking advantage of the associations between compounds as they are created in biochemical reactions. This suggests that the nervous system can classify odors based on statistics of their co-occurrence within natural mixtures rather than from the chemical structures of the ligands themselves. We show that this statistical perspective makes it possible to map odors to points in a hyperbolic space. Hyperbolic coordinates have a long but often underappreciated history of relevance to biology. For example, these coordinates approximate the distance between species computed along dendrograms and, more generally, between points within hierarchical tree-like networks. We find that both natural odors and human perceptual descriptions of smells can be described using a three-dimensional hyperbolic space. This match in geometries can avoid distortions that would otherwise arise when mapping odors to perception.

Predicting human olfactory perception from chemical features of odor molecules

It is still not possible to predict whether a given molecule will have a perceived odor or what olfactory percept it will produce. We therefore organized the crowd-sourced DREAM Olfaction Prediction Challenge. Using a large olfactory psychophysical data set, teams developed machine-learning algorithms to predict sensory attributes of molecules based on their chemoinformatic features. The resulting models accurately predicted odor intensity and pleasantness and also successfully predicted 8 among 19 rated semantic descriptors ("garlic," "fish," "sweet," "fruit," "burnt," "spices," "flower," and "sour"). Regularized linear models performed nearly as well as random forest-based ones, with a predictive accuracy that closely approaches a key theoretical limit. These models help to predict the perceptual qualities of virtually any molecule with high accuracy and also reverse-engineer the smell of a molecule.

A Mathematical Model of the Olfactory Bulb for the Selective Adaptation Mechanism in the Rodent Olfactory System

To predict the odor quality of an odorant mixture, the interaction between odorants must be taken into account. Previously, an experiment in which mice discriminated between odorant mixtures identified a selective adaptation mechanism in the olfactory system. This paper proposes an olfactory model for odorant mixtures that can account for selective adaptation in terms of neural activity. The proposed model uses the spatial activity pattern of the mitral layer obtained from model simulations to predict the perceptual similarity between odors. Measured glomerular activity patterns are used as input to the model. The neural interaction between mitral cells and granular cells is then simulated, and a dissimilarity index between odors is defined using the activity patterns of the mitral layer. An odor set composed of three odorants is used to test the ability of the model. Simulations are performed based on the odor discrimination experiment on mice. As a result, we observe that part of the neural activity in the glomerular layer is enhanced in the mitral layer, whereas another part is suppressed. We find that the dissimilarity index strongly correlates with the odor discrimination rate of mice: r = 0.88 (p = 0.019). We conclude that our model has the ability to predict the perceptual similarity of odorant mixtures. In addition, the model also accounts for selective adaptation via the odor discrimination rate, and the enhancement and inhibition in the mitral layer may be related to this selective adaptation.

Olfactory perception of chemically diverse molecules

Background

Understanding the relationship between a stimulus and how it is perceived reveals fundamental principles about the mechanisms of sensory perception. While this stimulus-percept problem is mostly understood for color vision and tone perception, it is not currently possible to predict how a given molecule smells. While there has been some progress in predicting the pleasantness and intensity of an odorant, perceptual data for a larger number of diverse molecules are needed to improve current predictions. Towards this goal, we tested the olfactory perception of 480 structurally and perceptually diverse molecules at two concentrations using a panel of 55 healthy human subjects.

Results

For each stimulus, we collected data on perceived intensity, pleasantness, and familiarity. In addition, subjects were asked to apply 20 semantic odor quality descriptors to these stimuli, and were offered the option to describe the smell in their own words. Using this dataset, we replicated several previous correlations between molecular features of the stimulus and olfactory perception. The number of sulfur atoms in a molecule was correlated with the odor quality descriptors “garlic,” “fish,” and “decayed,” and large and structurally complex molecules were perceived to be more pleasant. We discovered a number of correlations in intensity perception between molecules. We show that familiarity had a strong effect on the ability of subjects to describe a smell. Many subjects used commercial products to describe familiar odorants, highlighting the role of prior experience in verbal reports of olfactory perception. Nonspecific descriptors like “chemical” were applied frequently to unfamiliar odorants, and unfamiliar odorants were generally rated as neither pleasant nor unpleasant.

Conclusions

We present a very large psychophysical dataset and use this to correlate molecular features of a stimulus to olfactory percept. Our work reveals robust correlations between molecular features and perceptual qualities, and highlights the dominant role of familiarity and experience in assigning verbal descriptors to odorants.

Electronic supplementary material

The online version of this article (doi:10.1186/s12868-016-0287-2) contains supplementary material, which is available to authorized users.

Odorant receptors of Drosophila are sensitive to the molecular volume of odorants

Which properties of a molecule define its odor? This is a basic yet unanswered question regarding the olfactory system. The olfactory system of Drosophila has a repertoire of approximately 60 odorant receptors. Molecules bind to odorant receptors with different affinities and activate them with different efficacies, thus providing a combinatorial code that identifies odorants. We hypothesized that the binding affinity of an odorant-receptor pair is affected by their relative sizes. The maximum affinity can be attained when the molecular volume of an odorant matches the volume of the binding pocket. The affinity drops to zero when the sizes are too different, thus obscuring the effects of other molecular properties. We developed a mathematical formulation of this hypothesis and verified it using Drosophila data. We also predicted the volume and structural flexibility of the binding site of each odorant receptor; these features significantly differ between odorant receptors. The differences in the volumes and structural flexibilities of different odorant receptor binding sites may explain the difference in the scents of similar molecules with different sizes.