On March 21, the latest research by Chinese scientists was published in Molecular Plant. The Institute of Crop Sciences of the Chinese Academy of Agricultural Sciences, the National South Subtropical Agricultural Research Center and the Alibaba DAMO Academy (Lakeside Laboratory) jointly developed a "full-process intelligent breeding platform" that will revolutionize traditional breeding. The platform integrates the entire breeding process of breeding data management, analysis, calculation acceleration and AI prediction of parents and excellent varieties. At present, 23 scientific research institutions across the country have been drawn to "try it out".
Research paper information Source: Molecular Plant
Traditional breeding methods are time-consuming and costly, and often take ten years or more. An intelligent breeding platform is expected to greatly shorten the breeding cycle. The platform not only integrates modules such as germplasm resource information management, breeding experiment design and simulation, and high-throughput data analysis, but also expands the capacity for breeding data management and analysis. It is reported that the platform's data capacity, running speed, and data security have all reached the world's advanced level.
Intelligent breeding requires high-throughput data analysis, relies on massive computing and modeling, and complex operations, which poses a great challenge. The aforementioned units gave full play to their respective advantages and worked together to develop this fully functional and user-friendly full-process intelligent breeding platform. As a high-throughput computing platform, it can provide one-stop, low-threshold scientific support services for breeders on the web. At present, 23 scientific research institutions around the world have taken the lead in adopting the platform, including the China National Rice Research Institute, the Institute of Genetics and Developmental Biology of the Chinese Academy of Sciences, the International Livestock Research Institute, and Wageningen University in the Netherlands.
Main modules of the smart breeding platform Provided by DAMO Academy
To accelerate the calculation of breeding decisions, the smart breeding platform introduces a variety of big data computing, artificial intelligence models, and automated processing modules. Gu Fei, the corresponding author of the paper and a smart breeding scientist at DAMO Academy, introduced that the platform realizes 110 times acceleration of variant site calculation for gene sequencing data, more than 25 times acceleration of genotype filtering, and more than 1000 times acceleration of population genetics analysis. For the whole genome selection algorithm, the platform's accuracy is improved by an average of 15%, and the prediction stability is improved by 5%. In addition, the platform's data storage capacity is up to 1PB, which can support hundreds of users to carry out efficient breeding big data management, scheduling, and analysis at the same time, realizing the acceleration of breeding decision calculation.
It is also introduced that the construction of the smart breeding platform started in June 2021, and was jointly constructed by Alibaba Charity Foundation, DAMO Academy, and many other units such as the Crop Research Institute of Chinese Academy of Agricultural Sciences.
Related paper information:
https://www.cell.com/molecular-plant/fulltext/S1674-2052(24)00075-3
