ADDO: a comprehensive toolkit to detect, classify and visualize additive and non-additive quantitative trait loci
ADDO: a comprehensive toolkit to detect, classify and visualize additive and non-additive quantitative trait loci
复制标题
DOI：
10.1093/bioinformatics/btz786
复制
发表时间：
2020-03-01
期刊：
影响因子：
5.8
通讯作者：
Huang, Lusheng
中科院分区：
文献类型：
--
作者：
Cui, Leilei;Yang, Bin;Huang, Lusheng
动机：在过去的十年中，全基因组关联研究（GWAS）已被用来绘制复杂性状背后的数量性状基因座（QTL）图谱。然而，大多数 GWAS 侧重于加性遗传效应，而忽略非加性效应，假设大多数 QTL 具有加性作用。因此，由显性和其他非加性效应驱动的 QTL 可能会被忽略。结果：我们开发了 ADDO，这是一种高效的工具，用于检测、分类和可视化具有加性和非加性效应的 QTL。 ADDO 实施混合模型转换来控制群体结构和不平等相关性，从而解释个体之间的加性和显性遗传协方差，并将单核苷酸多态性效应分解为加性、部分显性、显性或过度显性。矩阵乘法方法用于加速计算：使用 10 个 CPU 对 900 个人的 1300 万个标记进行基因组扫描大约需要 5 小时。模拟数据分析证实了 ADDO 在具有不同加性和显性遗传方差分量的性状上的性能。我们展示了远交大鼠的两个真实例子，其中 ADDO 识别了加性模型无法检测到的显着显性 QTL。 ADDO提供了一个系统的管道来表征全基因组序列数据中的加性和非加性QTL，它补充了当前主流的加性遗传效应GWAS软件。
Motivation: During the past decade, genome-wide association studies (GWAS) have been used to map quantitative trait loci (QTLs) underlying complex traits. However, most GWAS focus on additive genetic effects while ignoring non-additive effects, on the assumption that most QTL act additively. Consequently, QTLs driven by dominance and other non-additive effects could be overlooked.Results: We developed ADDO, a highly efficient tool to detect, classify and visualize QTLs with additive and non-additive effects. ADDO implements a mixed-model transformation to control for population structure and unequal relatedness that accounts for both additive and dominant genetic covariance among individuals, and decomposes single-nucleotide polymorphism effects as either additive, partial dominant, dominant or over-dominant. A matrix multiplication approach is used to accelerate the computation: a genome scan on 13 million markers from 900 individuals takes about 5 h with 10 CPUs. Analysis of simulated data confirms ADDO's performance on traits with different additive and dominance genetic variance components. We showed two real examples in outbred rat where ADDO identified significant dominant QTL that were not detectable by an additive model. ADDO provides a systematic pipeline to characterize additive and non-additive QTL in whole genome sequence data, which complements current mainstream GWAS software for additive genetic effects.