A mathematical model that maps the relationship between the molecular structure of odorous substances and how they are perceived is claimed to have an edge over human experts, when it comes to identifying smells correctly.
A central undertaking of neuroscience is learning how our senses translate light into sight, sound into hearing, food into taste, and texture into touch. Smell is where these sensory relationships get more complex and perplexing.
To address this question, a research team from the Monell Chemical Senses Centre and start-up firm Osmo – a spinout from machine learning work undertaken at Google Research – are investigating how airborne chemicals connect to odour perception in the brain. They report that a machine-learning model has achieved human-level proficiency at describing, in words, how chemicals might smell. The research appeared in the 1 September issue of Science.
“The model addresses age-old gaps in the scientific understanding of the sense of smell,” said senior co-author Joel Mainland.
This collaboration moves the world closer to digitizing odours, which can be recorded and reproduced. It also may identify new odours for the fragrance and flavour industry that could decrease dependence on endangered plants, and identify new functional scents for things like mosquito repellent.
Humans have about 400 functional olfactory receptors. These are proteins at the end of olfactory nerves that connect with airborne molecules to transmit an electrical signal to the olfactory bulb. The number of olfactory receptors is much more than we use for colour vision – four – or even taste – about 40.
“In olfaction research, however, the question of what physical properties make an airborne molecule smell the way it does to the brain has remained an enigma,” said Mainland. “But if a computer can discern the relationship between how molecules are shaped and how we ultimately perceive their odours, scientists could use that knowledge to advance the understanding of how our brains and noses work together.”
To address this, Osmo created a model that learned how to match the prose descriptions of a molecule’s odour with molecular structure. The resulting map of these interactions is essentially groupings of similarly smelling odours, like floral and candy sweets. “Computers have been able to digitize vision and hearing, but not smell – our deepest and oldest sense,” said Wiltschko. “This study proposes and validates a novel data-driven map of human olfaction, matching chemical structure to odour perception.”
What is the smell of garlic or of ozone?
The model was trained using an industry dataset that included the molecular structures and odour qualities of 5,000 known odourants. Data input is the shape of a molecule, and the output is a prediction of which odour words best describe its smell.
To ascertain the efficacy of the model, researchers conducted a blind validation procedure in which a panel of trained participants described new molecules, and then compared their answers with the model’s description. The 15 panelists were each given 400 odourants as well as trained to use a set of 55 words – from mint to musty – to describe each molecule.
In comparing the model’s performance to that of individual panelists, the model achieved better predictions of the average of the group’s odour ratings than any single panelist in the study, impurities aside. Specifically, the model performed better than the average panelist for 53% of the molecules tested.
“The most surprising result, however, is that the model succeeded at olfactory tasks it was not trained to do,” said Mainland. “The eye-opener was that we never trained it to learn odour strength, but it could nonetheless make accurate predictions.”
The model was able to identify dozens of pairs of structurally dissimilar molecules that had counter-intuitively similar smells, and characterize a wide variety of odour properties, such as odour strength, for 500,000 potential scent molecules.