An exponential dispersion model for the distribution of human single nucleotide polymorphisms
An exponential dispersion model for the distribution of human single nucleotide polymorphisms
复制标题
DOI：
10.1093/molbev/msg057
复制
发表时间：
2003-04-01
影响因子：
10.7
通讯作者：
Kendal, WS
中科院分区：
文献类型：
--
作者：
Kendal, WS
对国际SNP图谱工作组绘制的142万个人类单核苷酸多态(SNPs)的分析表明，估计方差与每个样本库的SNPs平均数之间存在明显的幂函数关系。这种关系可以通过假设尺度不变的泊松伽马(PG)指数分布模型来描述SNPs在垃圾桶中的分布来解释。在这个模型中，样本箱将包含随机(泊松分布)数量相同的下降基因组片段，每个片段都有独立分布和伽马分布数量的SNPs。该模型与传统的聚结模型在定性和定量上都是一致的。它与从SNP图得到的经验累积分布函数以及模拟数据相一致。该模型被用来估计每条染色体的杂合度，以及单倍型区块的平均数量和大小。这些估计与传统研究的测量结果一致。因此，该PG模型为描述SNPs分布提供了一种替代蒙特卡罗模拟的方法。
An analysis of 1.42 million human single nucleotide polymorphisms (SNPs), mapped by the International SNP Map Working Group, revealed an apparent power function relationship between the estimated variance and mean number of SNPs per sample bin. This relationship could be explained by the assumption that a scale invariant Poisson gamma (PG) exponential dispersion model could describe the distribution of SNPs within the bins. In this model the sample bins would contain random (Poisson distributed) numbers of identical by descent genomic segments, each with independently distributed and gamma distributed numbers of SNPs. This model was both qualitatively and quantitatively consistent with the conventional coalescent model. It agreed with the empirical cumulative distribution functions derived from the SNP maps as well as with simulated data. The model was used to estimate the heterozygosity it, and the mean number and size of haplotype blocks for each chromosome. These estimates were consistent with measurements from conventional studies. This PG model thus provides an alternative to Monte Carlo simulation for description of the distribution of SNPs.