剖析复杂性状遗传规律，是当代生命科学研究的热点，也面临巨大的挑战。目前基于全基因组测序信息的关联分析和基因定位，无法有效检测上位性基因和与环境互作的基因。本项目将构建可以综合分析作物育种群体多年多点实验数据的基因定位遗传模型，包括加性、显性、上位性等遗传效应和基因与环境的互作效应（加性×环境、显性×环境、上位性×环境），研制基于混合线性模型的统计分析方法。提出对作物纯系群体（品种和种质资源材料）和双列杂交群体（亲本及其F1组合）复杂性状进行全基因组SNP定位和转录组定位的新方法。检测陆地棉纯系品种高产、优质性状的关键SNP位点和转录位点，筛选陆地棉强优势组合的关键SNP位点和转录位点。提出基于全基因组QTSs和QTTs基因功能预测和分子选择进展预测的分析工具和育种策略。

Most agronomic traits of crops are complex traits, which are controlled by a set of genes with gene-by-gene interaction and gene-by-environment interaction. Due to the advanced high-throughput biological technologies, it is now convenient to acquire large-scale molecular data of SNPs and transcripts. It brings challenges for dissecting the genetic architecture of complex traits. We are going to develop new genetic models for mapping genes with additive, dominance and epistasis effects, as well as their environment interactions. We will also propose statistical methodologies for deliver unbiased estimation for gene locations and their genetic effects. The efficiency and effectiveness of the proposed genetic models and statistical methods will be investigated by Monte Carlo simulations and real data analysis. A diallel cross mating design will be applied to 150 parents of Upland cotton for producing 200 F1 crosses. These genotypes will be grown in three locations for two years, and will be conducted genotyping by whole-genome SNPs and transcripts, and also observing agronomic traits. GWAS will be applied for mapping quantitative trait SNPs (QTSs) and quantitative trait transcripts (QTTs) of agronomic traits in cotton.

作物复杂农艺性状的分子育种改良已经进入基于基因组、转录组、蛋白组和代谢组的精准分子选择的新阶段。本项目应用我们在上一个国家自然科学基金（3047091）资助下开发的基于多组学高通量表达的基因型变异与复杂性状关联分析的新方法和软件QTXNetwok，在分子水平上剖析了基因组和转录组调控作物重要农艺性状的遗传规律，提出了基于SNPs和转录子实现精准分子选择遗传改良的途径。并进一步把所研制的新方法成功应用于对人类和动物复杂性状的遗传剖析和分子诊断。