Could you use AI to predict what a chemical compound smells like?

The model is very good at things like garlic and fishy.

It's much worse at things like Musk.

It's Wednesday, November 29th, and it's also Science Friday.

I'm Scifry producer D. Peter Schmidt. If you want to predict the color of a certain kind of light or what something sounds like,

you can use wavelength or frequencies to get a good idea of what you'll see or hear.

But predicting smells is more difficult.

Chemical compounds can look quite similar to each other but produce completely different smells.

But scientists are starting to use AI models to make more accurate smell predictions.

Aerofleito talks to a researcher about how this new technique could

advance smell science. Researchers describe developing an AI model that if you give it the structure of

a chemical compound can predict where it's likely to fall on a map of odors, for instance,

is it more grassy, more meaty, more floral?

Dr. Joel Mainland is one of the authors of that report. He's a member of the Monel Chemical

Census Center and an adjunct associate professor in the Department of Neuroscience at the University of Pennsylvania in Philadelphia.

Welcome to Science Friday.

Thanks for having me.

Nice to have you.

Okay, before we start on this new work, let's get a refresher on Smell Biology 101.

I was always taught that it's sort of a lock and a key situation in your nose with

smell molecules fitting into like a lock receptors receptors is that right?

Yeah, so I think that's a good analogy at the basic level and gets most of the things we care about correct.

...