Artificial Intelligence drives 'computerized smell' for insect control

Machine smell startup Osmo was established in January 2023 and received US$60 million in Series A funding led by Lux Capital and Google Ventures. Osmo combines machine learning, data science, psychophysics, olfactory neuroscience, electrical engineering and chemistry in a multidisciplinary approach to digitizing smell.

Osmo’s work is based on machine smell research validated by the Google Research team, including a 2019 study using graph neural network predictions The smell of molecules. The company has begun testing the waters in the fragrance market, aiming to create a new generation of better, safer and more environmentally friendly fragrance molecules. Over time, Osmo hopes to achieve greater success in areas such as public health and agriculture

Recently, Osmo announced it received a $350 grant from the Bill and Melinda Gates Foundation A $10,000 grant to advance the company’s AI-powered scent platform to discover and produce compounds that repel, attract, or destroy disease-carrying insects to improve animal and human health. The funding complements the $5 million equity investment the Gates Foundation made in Osmo when it was founded in January 2023.

The World Health Organization (WHO) estimates that mosquitoes worldwide Disease-carrying insects kill millions of people. Since insects rely heavily on their sense of smell to navigate and locate potential targets, scent becomes the most direct way to guide disease-carrying insects away from humans. Developing compounds with specific odors that can effectively repel or deter insects, thereby disrupting their attraction to human hosts, minimizing disease transmission and providing a targeted, effective method of insect control

Osmo CEO Alex Wiltschko said: “By using new odor molecules, we can more effectively guide disease-carrying insects away from human contact, potentially saving millions of lives.” Out of a vast space of billions of potential molecules, only a few thousand have been screened for this capability. With generous support from the Gates Foundation, we are using our AI-powered scent platform to analyze this vast universe of chemistry. to analyze the space and discover novel agents capable of changing insect behavior to prevent disease that are effective, safe and affordable for both human and animal health."

This The grant follows a proof-of-concept pilot previously funded by the foundation and will be published as a research paper in late 2022. In the pilot, the research team trained a state-of-the-art computational model on the largest mosquito repellent dataset in history to date. The team experimentally evaluated the model on about 400 repellent molecules of different chemical properties, identifying eight that were more repellent than the widely used DEET and Picaridin. of molecules.

In the current project, Osmo will build on a proof-of-concept pilot with the main goals of:

By incorporating more data and more testing of more

compounds, the Previous studies expanded this by at least tenfold.

Using machine learning technology, we can discover those promising, novel, cheap, and diverse chemical candidate molecules

We need to develop Predictive models to take into account real-world constraints associated with candidate molecules, including cost, spatial range, biodegradability, toxicity and human odor perception

We need to synthesize, test and Optimizing novel repellent candidates for human trials and ecological impact assessment

We can use this repellent model to demonstrate the effectiveness of machine learning in discovering new mosquito attractants, This attractant will be superior to existing attractants

Osmo's insect control platform design uses the latest machine learning and generative artificial intelligence technology, which allows the system to Evaluate the potential effectiveness and safety of billions of molecules in just seconds

" New machine learning The method has important potential to accelerate the discovery and design of improved mosquito repellents and attractants. "Osmo's model shows great promise, and I'm excited about the team's progress in the coming years." ”

Osmo’s insect control work is part of the company’s mission to improve the health and well-being of human life by giving computers a sense of smell. The core of this mission is built by the Osmo team An "odor map" for predicting the structure and odor of molecules

"Osmo reveals surprising connections between insect and human olfaction, and our odor map predicts the olfactory effects of molecules on humans and insects," Wiltschko said. "Our mission to digitize olfaction will have many potential ways to make the world a healthier, happier place. We are all particularly excited about using our maps to design entirely new molecules to stop the spread of insect-borne diseases."

The above is the detailed content of Artificial Intelligence drives 'computerized smell' for insect control. For more information, please follow other related articles on the PHP Chinese website!