Selection and breeding of bilberry varieties with high anthocyanin content and supporting light cultivation methods

Abstract

The application relates to the technical field of raspberry breeding and cultivation, in particular to a breeding and matching light and simplified cultivation method of raspberry varieties with high anthocyanin content, which comprises the following steps: screening core effect sites related to anthocyanin synthesis by using genome-wide association study; Deconstruct the feature space to obtain the phenological metabolic characteristics, train the growth stage recognizer and construct the metabolic pathway switching gating model; Based on phenological attribution probability, a multi-subset expert model for environmental signal regulation is constructed; Input real-time habitat data into gated model, and perform single expert analysis or multi-expert model collaborative weighted fusion according to fitness probability to obtain agronomic parameters and convert them into anthocyanin content prediction values. The application can obviously improve the accuracy of anthocyanin accumulation and the stability of variety breeding of raspberry, and provide accurate decision support for high-quality breeding and efficient cultivation.