This is the authors' abstract. We don't add key points for this paper.
In real data analysis, most commonly used genome-wide association study (GWAS) methods often miss some important loci and trait heritability. To address these challenges, Li et al (2022a) established an innovative method named 3VmrMLM based on a compressed variance component mixed model. In 3VmrMLM, all the effects in quantitative trait nucleotide (QTN), QTN-by-environment interaction (QEI), and QTN-by-QTN interaction (QQI) detection are compressed into an effect-related vector, while all polygenic backgrounds are compressed into a vector-related polygenic background. This method is especially well-suited for species with a high proportion of heterozygous genotypes, such as human, forests, chrysanthemums, and grasslands. Can 3VmrMLM replace existing methods? The answer is no, despite 3VmrMLM demonstrating superiority over existing methods. For the detection of loci dominated by additive effects, existing methods remain appropriate, as observed in rice, wheat, and soybean.Since GWAS is based on linkage disequilibrium from historical recombination, there is complementarity between methods (Zhang et al., 2019). However, existing methods face challenges in detecting dominant effects and small allele substitution effects (Zhang et al., 2023).When analyzing real data, the inflation factor or quantile-quantile plot serves as a common metric to assess method performance. However, this is not crucial for our mrMLM and 3VmrMLM methods (Zhang et al., 2020;Li et al., 2022a), because their genome-wide scanning aims to select potentially associated markers rather than identify significant loci. A method is considered effective when it mines some importantly known and candidate genes around these loci, supported by strong evidence, as seen in 3VmrMLM. These identified loci may be used for genomic selection (Su et al., 2024), while more associated known and candidate genes can be mined and highlighted in the Manhattan plot. This Topic contains three articles focusing on methodological studies and comparisons. Yang Yield is one of the paramount breeding objectives, with nine articles in the Topic focusing on identifying QTNs and/or QEIs for yield-related traits. Zhang to perform association analysis between three measured grain size traits of 159 rice accessions in two environments and 2,017,495 SNPs, identifying 393 QTNs and 8 QEIs. They found 22 genes around QTNs and two genes around QEIs to be genuinely associated with these traits.Additionally, 14 candidate genes were significant in differential expression, GO annotation, and haplotype analysis. Moreover, in a joint analysis of main crop and ratoon rice, four known genes, eight additional candidate genes, and two candidate gene-by-environment interactions (GEIs) were identified as responsible for grain size-related traits. well-watered, drought, and heat stress conditions. They reported 34 genes from previous studies, confirming genes associated with drought tolerance (ereb53 and thx12) and heat stress (hsftf27 and myb60). Differential expression, tissue-specific expression, and haplotype analysis confirmed 24 candidate genes, while three yield GEIs (GRMZM2G064159, GRMZM2G146192, and GRMZM2G114789) were predicted. To effectively identify QEIs across diverse environments and QQIs across numerous markers, it is imperative to devise new algorithms tailored to sample size, computational speed and minimal memory requirements to meet the needs of human large data analysis. As the field advances, the genetic model for quantitative traits may transition from the classic Fisher genetic model to more comprehensive framework through the integration of artificial intelligence. We anticipate that our compressed variance component mixed model will emerge as a pivotal tool in the genetic analysis of complex traits and multi-omics data in the future.YMZ wrote the draft. All the authors revised, checked, and confirmed the final version.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: