Tree Genetics and Molecular Breeding 2024, Vol.14, No.2, 95-105 http://genbreedpublisher.com/index.php/tgmb 99 These methods help researchers understand the genetic relatedness and population distribution of samples. By identifying and correcting genetic background differences within the sample population, they reduce the occurrence of false-positive associations, providing a reliable basis for interpreting and understanding GWAS results. 3.2 Association analysis Association analysis is the core step of GWAS, aiming to detect statistical associations between genotypes and phenotypes. Through association analysis, researchers can determine whether a significant correlation exists between genotypes and phenotypes, thereby identifying genetic markers associated with specific traits. In association analysis, commonly used statistical models include linear mixed models (LMM) and generalized linear mixed models (GLMM), which can account for complex factors such as population structure and kinship relationships, reducing the occurrence of false-positive associations and improving the reliability and accuracy of analysis results (Alamin et al., 2022). In the association analysis process, appropriate statistical models need to be constructed to integrate genotype and phenotype data. Then, statistical methods are used to evaluate the degree of association between genotypes and phenotypes and determine which genetic markers are significantly associated with specific traits. Finally, by performing statistical significance tests and multiple testing corrections on the association results, the most promising candidate markers can be selected. In addition to considering population structure and kinship relationships, association analysis must also account for other factors that may influence the analysis results, such as environmental factors and phenotypic measurement errors. By comprehensively considering these factors, association analysis can more accurately reveal the relationships between genotypes and phenotypes, providing important data support for genetic research on ornamental trees. 3.3 Gene localization and functional prediction Once genetic markers significantly associated with a specific trait are discovered, the next crucial step is to localize the genes near these markers and predict their functions. This process is called gene localization and functional prediction, an important step in GWAS research (Mancuso et al., 2017). Through GWAS analysis, significantly associated genetic markers, i.e., associated SNPs, can be identified for a specific trait. The next task is to determine the genes near these SNPs, which can be achieved by searching for genes in the genome that are adjacent to the SNP positions. Generally, genes located near the associated SNP are considered potentially related to the target trait. After identifying the genes near the associated SNP, the next step is to predict their functions. This process involves using bioinformatics tools and databases to annotate and infer the functions of candidate genes in the associated region. Common functional prediction methods include genome sequence alignment, protein structure prediction, and functional domain analysis. Through these methods, information can be inferred about the biological processes, molecular functions, and cellular functions potentially involved by the candidate genes. Through gene localization and functional prediction, we can gain a deeper understanding of the genes associated with specific traits and their functions, providing important clues for further functional validation and applied research. This information helps guide the selection and breeding of ornamental trees, as well as the improvement and optimization of tree traits. 3.4 Experimental validation Ideally, the results of GWAS research need to be validated through experimental methods to confirm the true relationship between the identified genes and specific traits. This validation typically involves experimental proof of gene function and can employ various experimental methods. One common validation method is to use gene editing techniques, such as CRISPR-Cas9, to perform gene knockout or overexpression experiments on candidate genes. By conducting these experiments in model plants, direct observations can be made on the functional changes of the target genes, confirming their role in the
RkJQdWJsaXNoZXIy MjQ4ODYzMg==