"PlantRNA-FM, a groundbreaking AI model developed by a collaborative effort between the John Innes Centre and the University of Exeter, is designed to decode the genetic “language” of plants. Trained on RNA data from over 1,100 plant species, this innovative model analyzes genetic patterns to accelerate advancements in plant science, improve crop yields, and address global agricultural challenges.
This first-of-its-kind AI model offers a major technological breakthrough, enabling new discoveries and innovations in plant research, as well as potentially in the study of invertebrates and bacteria. RNA, like DNA, is crucial for carrying genetic information in living organisms. Its sequences and structural patterns serve as a genetic alphabet, similar to how words and phrases are constructed in human language.
To unravel the complexities of RNA’s structure, Professor Yiliang Ding’s research group at the John Innes Centre, known for studying RNA folding and its role in biological functions like plant growth and stress response, collaborated with Dr. Ke Li’s team at the University of Exeter. Their combined efforts led to the development of PlantRNA-FM, trained on an extensive dataset containing 54 billion RNA data points from 1,124 plant species."
"When developing PlantRNA-FM, the researchers used a similar approach to training AI models like ChatGPT, which are designed to understand human language. PlantRNA-FM was trained by analyzing RNA data from plant species around the world, giving it an in-depth understanding of RNA's role across the plant kingdom.
Much like ChatGPT comprehends and generates human language, PlantRNA-FM has learned to interpret the structure and logic of RNA sequences. The model has already made accurate predictions about RNA functions and identified specific RNA structural patterns within plant transcriptomes. These predictions have been validated by experiments, confirming that the RNA structures identified by the model influence how effectively genetic information is translated into proteins.
"Although RNA sequences may seem random to the naked eye, our AI model has uncovered hidden patterns within them," explains Dr. Haopeng Yu, a postdoctoral researcher at the John Innes Centre.
This success is the result of a collaborative effort, which also involved scientists from Northeast Normal University and the Chinese Academy of Sciences in China.
Professor Yiliang Ding commented, "PlantRNA-FM is just the beginning. Together with Dr. Li’s group, we are advancing AI techniques to further decode the hidden languages of DNA and RNA in nature. This breakthrough paves the way for deeper insights into plants and may revolutionize crop improvement and AI-based gene design. AI is increasingly pivotal in helping scientists tackle global challenges, such as feeding a growing population and developing crops that can adapt to climate change."
