Scientists Successfully Constructed an Intelligent RNA Structure Prediction Model

Recently, scientists from Stanford University in the United States have successfully developed an RNA structure prediction model - ARES. This model utilizes artificial intelligence technology to make accurate predictions of the three-dimensional structure of RNA through autonomous learning of a small amount of data. The relevant research results were published in Science under the title of Geometric deep learning of RNA structure.

RNA, as an important regulator of life activities, can not only determine the protein content, but also transmit rich cellular activity information. The three-dimensional structure of RNA is crucial for its related functions, but due to its fast degradation rate, it is difficult to determine its structure through conventional experimental methods. This study integrates cutting-edge artificial intelligence achievements such as neural network technology to construct an RNA prediction model with significantly better accuracy than traditional tools. Through short-term learning, intelligent predictions of RNA structure can be made. Researchers trained an ARES model that can accurately predict using RNA data from 18 known structures.

This research achievement provides a new method for accurately predicting the three-dimensional structure of RNA and developing new targets for RNA type drugs.

Prev: Scientists explore the use of Bessel beams to improve porosity and defect issues in 3D metal printing