Researchers in Shenzhen, China, led by Zhou Yongfeng from the Agricultural Genomics Institute at the Chinese Academy of Agricultural Sciences, have developed a groundbreaking method for grape breeding. This method, leveraging artificial intelligence (AI), has the potential to reduce the grape breeding cycle significantly, with a prediction accuracy of 85 percent, effectively quadrupling breeding efficiency compared to traditional practices.
Zhou Yongfeng stated, "This research is expected to achieve precise breeding design for grapes, accelerate grape variety innovation, and provide a reference for breeding other perennial crops." The findings were documented in the journal Nature Genetics.
The team's work, initiated in 2015, has culminated in the creation of the first complete grape genome map and the first grape pan-genome, Grapepan v1.0, in 2023. To explore the link between grape genes and characteristics, over 400 grape varieties were analyzed for traits such as cluster size and berry color, leading to the construction of genetic and trait maps for grapes.
To utilize the genomic data for breeding, the research team employed machine learning to develop a predictive model. This model facilitates early prediction and selection of grape varieties, optimizing breeding strategies. "With this model, breeders can assess the genetic potential of a large number of breeding materials quickly and accurately, enabling them to better select superior varieties," Zhou explained.
The research has led to the approval of six national invention patents and the filing of one international patent application, marking a significant advancement in the field of agricultural genomics.
Source: XINHUANET