Shaw S H, Carrasquillo M M, Kashuk C, Puffenberger E G, Chakravarti A
Department of Genetics and Center for Human Genetics, Case Western Reserve University School of Medicine and University Hospitals of Cleveland, Cleveland, Ohio 44106 USA.
Genome Res. 1998 Feb;8(2):111-23. doi: 10.1101/gr.8.2.111.
Genetic studies of complex hereditary disorders require for their mapping the determination of genotypes at several hundred polymorphic loci in several hundred families. Because only a minority of markers are expected to show linkage and association in family data, a simple screen of genetic markers to identify those showing linkage in pooled DNA samples can greatly facilitate gene identification. All studies involving pooled DNA samples require the comparison of allele frequencies in appropriate family samples and subsamples. We have tested the accuracy of allele frequency estimates, in various DNA samples, by pooling DNA from multiple individuals prior to PCR amplification. We have used the ABI 377 automated DNA sequencer and GENESCAN software for quantifying total amplification using a 5' fluorescently labeled forward PCR primer and relative peak heights to estimate allele frequencies in pooled DNA samples. In these studies, we have genotyped 11 microsatellite markers in two separate DNA pools, and an additional four markers in a third DNA pool, and compared the estimated allele frequencies with those determined by direct genotyping. In addition, we have evaluated whether pooled DNA samples can be used to accurately assess allele frequencies on transmitted and untransmitted chromosomes, in a collection of families for fine-structure gene mapping using allelic association. Our studies show that accurate, quantitative data on allele frequencies, suitable for identifying markers for complex disorders, can be identified from pooled DNA samples. This approach, being independent of the number of samples comprising a pool, promises to drastically reduce the labor and cost of genotyping in the initial identification of disease loci. Additional applications of DNA pooling are discussed. These developments suggest that new statistical methods for analyzing pooled DNA data are required.
复杂遗传性疾病的基因研究需要在数百个家庭中确定数百个多态性位点的基因型，以便进行图谱绘制。由于在家族数据中预计只有少数标记会显示连锁和关联，因此通过简单筛选遗传标记以识别在混合DNA样本中显示连锁的标记，可以极大地促进基因鉴定。所有涉及混合DNA样本的研究都需要比较适当的家族样本和子样本中的等位基因频率。我们通过在PCR扩增前将多个个体的DNA混合，测试了各种DNA样本中等位基因频率估计的准确性。我们使用ABI 377自动DNA测序仪和GENESCAN软件，使用5'荧光标记的正向PCR引物对总扩增进行定量，并通过相对峰高估计混合DNA样本中的等位基因频率。在这些研究中，我们对两个单独的DNA池中11个微卫星标记进行了基因分型，并对第三个DNA池中的另外4个标记进行了基因分型，并将估计的等位基因频率与直接基因分型确定的频率进行了比较。此外，我们评估了在一系列家庭中使用等位基因关联进行精细结构基因图谱绘制时，混合DNA样本是否可用于准确评估传递和未传递染色体上的等位基因频率。我们的研究表明，可以从混合DNA样本中识别出准确的、定量的等位基因频率数据，适用于识别复杂疾病的标记。这种方法独立于构成一个池的样本数量，有望在疾病基因座的初步鉴定中大幅降低基因分型的工作量和成本。文中还讨论了DNA混合的其他应用。这些进展表明需要新的统计方法来分析混合DNA数据。
