Association mapping of gene expression data, generated from transcriptome and proteome studies, provides a means of understanding the functional significance and trait association potential of candidate genes. In this study, we applied candidate gene association mapping to validate sugarcane genes, using data from the starch and sucrose metabolism pathway, transcriptome, and proteome. We performed multiplex PCR targeted amplicon sequencing of 109 candidate genes, using NGS technology. A range of statistical models, both single-locus and multi-locus, were compared for minimization of false positives in association mapping of four sugar-related traits with different heritability. The Fixed and random model Circulating Probability Unification model effectively suppressed false positives for both low- and high-heritability traits. We identified favorable alleles of the candidate genes involved in signalling and transcriptional regulation. The results will support genetic improvement of sugarcane and may help clarify the genetic architecture of sugar-related traits.

由转录组和蛋白质组研究产生的基因表达数据的关联作图提供了一种了解候选基因的功能意义和性状关联潜力的方法。在这项研究中，我们应用候选基因关联作图来验证甘蔗基因，使用来自淀粉和蔗糖代谢途径、转录组和蛋白质组的数据。我们使用 NGS 技术对 109 个候选基因进行了多重 PCR 靶向扩增子测序。比较了一系列统计模型，包括单位点和多位点，以最大限度地减少四种具有不同遗传力的糖相关性状的关联映射中的假阳性。固定和随机模型循环概率统一模型有效地抑制了低遗传力和高遗传力性状的误报。我们确定了参与信号传导和转录调控的候选基因的有利等位基因。该结果将支持甘蔗的遗传改良，并可能有助于阐明糖相关性状的遗传结构。