The Intergovernmental Panel on Climate Change (IPCC) has declared that removing carbon from the atmosphere is now essential to combat climate change and limit the rise in global temperatures. To support these efforts, Salk scientists are leveraging plants' natural ability to extract carbon dioxide from the air by optimizing plant root systems to store more carbon for longer periods of time. using abilities.
To engineer these climate-saving plants, scientists at Salk's Plant Utilization Initiative developed a sophisticated artificial intelligence (AI) software called SLEAP that tracks multiple characteristics of root growth. Using new research tools. Created by Salk Fellow Tarmo Pereira, his SLEAP was initially designed to track the movements of animals in the lab. Pereira is currently working with plant scientist and Salk colleague Professor Wolfgang Busch to apply SLEAP to plants.
in Research published in plant phenomicsBusch and Pereira are using SLEAP to study plant root phenotypes: how deep and wide a plant grows and how large its root system becomes; We have debuted a new protocol for analyzing other physical properties. By applying SLEAP to plants, researchers were able to establish the most extensive catalog of plant root system phenotypes to date.
Additionally, tracking these physical root system traits will help scientists find genes associated with those traits and whether multiple root traits are determined by the same gene or independently. help you discover what's going on. This allows the Salk team to determine which genes are most beneficial for plant design.
“This collaboration truly proves that Salk science is special and impactful,” Pereira said. “We're not just 'borrowing' from different disciplines; we're actually putting them on equal footing to create something greater than the sum of its parts.”
Before using SLEAP, tracking the physical characteristics of both plants and animals required significant effort and slowed the scientific process. If researchers want to analyze images of plants, they must manually flag which parts of the image are plants and which parts are not, frame by frame, part by part, and pixel by pixel. Only then can older AI models be applied to process the images and collect data about the structure of the plant.
SLEAP features a unique use of both computer vision (the ability of computers to understand images) and deep learning (an AI approach to training computers to learn and behave like the human brain). It's about being there. This combination allows researchers to process images without going pixel by pixel, instead skipping this intermediate, labor-intensive step and jumping directly from image input to defined plant features. can do.
“We have created a robust protocol, validated across multiple plant types, that reduces analysis time and human error, while focusing on accessibility and ease of use.and We did not have to make any changes to the actual SLEAP software,” said first author Elizabeth Berrigan, a bioinformatics analyst in the Bush lab.
Researchers developed a downloadable toolkit for SLEAP without changing SLEAP's baseline technology. sleep route (Available as open source software here). and sleep routeSLEAP can handle biological characteristics of the root system such as depth, mass, and growth angle.
The Salk team is sleep route Packages of various plants including model plant species as well as crops such as soybean, rice, canola, etc. Arabidopsis—A flowering weed belonging to the mustard family. After testing a variety of plants, we found that the new SLEAP-based method is 1.5x faster at annotating, 10x faster at training AI models, and 10x faster at predicting plant structure based on new data, all with comparable or better accuracy than existing ones. It turns out that it outperforms the method. From before.
In conjunction with large-scale genome sequencing efforts to resolve genotypic data for numerous crop varieties, extrapolation of phenotypic data such as the root systems of plants that grow deep in the soil is being conducted. You can understand the genes involved in the formation.
This step of linking phenotype and genotype is crucial in Salk's mission to create plants that retain more carbon for longer. This is because these plants require root systems that are designed to be deeper and stronger. By implementing this accurate and efficient software, the Harnessing Plants Initiative will be able to link desired phenotypes to target genes with breakthrough ease and speed.
“We have already been able to create the most extensive catalog of plant root system phenotypes to date, which will really accelerate research to create carbon-capturing plants to combat climate change. ,” said Bush, who holds the Hess Chair in Plant Science at Salk University. “Thanks to Talmo's professional software design, SLEAP was very easy to apply and use. It will become an essential tool in my laboratory in the future.”
SLEAP and sleep route.Because software and sleep route The toolkit is free to use, so researchers are looking forward to seeing how it is used. sleep route It will be used all over the world. The team has already begun discussions with NASA scientists, who hope to use the tool not only to help guide carbon-sequestering plants on Earth, but also to study plants in space.
At Salk, the joint team is not yet ready to disband. Already using SLEAP he has embarked on a new challenge: analyzing 3D data. Efforts to refine, expand, and share SLEAP sleep route Although it will continue for years to come, its use in Salk's Plant Utilization Initiative is already accelerating plant design and helping the institute impact climate change.
– This press release was originally published by salk website Edited for style and clarity
