Molecular Plant Breeding 2024, Vol.15, No.1, 15-26 http://genbreedpublisher.com/index.php/mpb 16 2 Technologies and Methods of Breeding 4.0 2.1 Genetic information integration The core of Breeding 4.0 lies in the integration of genetic information. By consolidating large-scale genetic data, such as genome sequencing, phenotype data, and environmental information, breeders can obtain comprehensive and accurate genetic backgrounds and trait expression, guiding the formulation of breeding strategies and the selection process. This integration of genetic information utilizes advanced genomics and data analysis technologies, such as genome-wide association study, expression profiling analysis, and machine learning algorithms, to comprehensively understand the genetic characteristics and potential trait expression of crops. In terms of genome-wide association study, the study “Genome Wide Association Study of Seedling and Adult Plant Leaf Rust Resistance in Elite Spring Wheat Breeding Lines” (PLoS One, 2016) applied the GWAS method to identify resistance genes or QTLs for leaf rust resistance in wheat. A total of 46 QTLs were identified, with approximately 20-30 contributing to field resistance to varying extents (Gao et al., 2016). Similarly, in the study “Genome-wide Association Study of Plant Architecture and Disease Resistance in Coffea canephora” (Euphytica, 2022), a genome-wide association analysis was conducted on plant architecture and disease resistance in Coffea canephora(de Faria Silva et al., 2022). In terms of expression profiling analysis, the study “Large-scale Transcriptional Profiling of Lignified Tissues in Tectona grandis” (BMC Plant Biol., 2015) conducted extensive transcriptome analysis of lignified tissues in teak (Tectona grandis), exploring expression differences across various developmental stages and transcription factors that potentially regulate lignin biosynthesis (Galeano et al., 2015). Similarly, the study “Genetic and Molecular Bases of Cucumber (Cucumis sativus L.) Sex Determination” (Mol. Breeding, 2019) delved into the genetic and molecular basis of sex determination in cucumber (Cucumis sativus L.), utilizing various methods including transcriptome analysis (Pawełkowicz et al., 2019). The application of machine learning algorithms in plant breeding has made significant progress. These algorithms can handle and analyze large-scale plant genotype and phenotype datasets, discovering meaningful patterns and thus driving the development of plant science and breeding. Technological advancements have enabled machine learning to play a crucial role in the analysis of various aspects (from biochemistry to yield) of plant phenotypes (van Dijk et al., 2021). Particularly in the field of image analysis, the application of deep learning technology, such as image segmentation, effectively reduces noise and bias, enhancing the accuracy of data analysis. Additionally, machine learning technology has shown tremendous potential in understanding the interactions and evolutionary patterns between plants and herbivores (Soltis et al., 2020). Furthermore, machine learning technologies have demonstrated their importance in handling high-dimensional multi-omics data, inferring gene regulatory networks, conducting correlation analysis of multi-omics data, and gene discovery. Modern machine learning technologies such as supervised learning, semi-supervised learning, unsupervised learning, and deep learning have been all applied in basic research in botany. In areas like plant phenomics, whole-genome selection-assisted breeding, genotype-to-phenotype prediction, and modeling the interaction between genotype and environment, machine learning technologies have also shown tremendous potential for application (Yan and Wang, 2022). 2.2 Genome editing Genome editing is another key technology in Breeding 4.0 (Xu et al., 2019). Through genome editing tools such as the CRISPR-Cas9 system, breeders can directly modify specific loci in the crop genome, achieving precise changes to target traits. Genome editing technology is characterized by its efficiency, precision, and repeatability, enabling effective modifications of single or multiple genes. This provides greater flexibility and accuracy in crop improvement. Bortesi and Fischer (2015) described CRISPR/Cas9 as a tool for specific double-strand DNA breaks, comparing it with other genome editing platforms such as ZFNs and TALENs. They emphasized its application and potential future development in plant breeding.
RkJQdWJsaXNoZXIy MjQ4ODYzMg==