Palta Priit, Kaplinski Lauris, Nagirnaja Liina, Veidenberg Andres, Möls Märt, Nelis Mari, Esko Tõnu, Metspalu Andres, Laan Maris, Remm Maido
Department of Bioinformatics, Institute of Molecular and Cell Biology, University of Tartu, Tartu, Estonia; Institute for Molecular Medicine Finland (FIMM), University of Helsinki, Helsinki, Finland.
Department of Bioinformatics, Institute of Molecular and Cell Biology, University of Tartu, Tartu, Estonia; Estonian Biocentre, Tartu, Estonia.
PLoS One. 2015 Apr 8;10(4):e0122713. doi: 10.1371/journal.pone.0122713. eCollection 2015.
DNA copy number variants (CNVs) that alter the copy number of a particular DNA segment in the genome play an important role in human phenotypic variability and disease susceptibility. A number of CNVs overlapping with genes have been shown to confer risk to a variety of human diseases thus highlighting the relevance of addressing the variability of CNVs at a higher resolution. So far, it has not been possible to deterministically infer the allelic composition of different haplotypes present within the CNV regions. We have developed a novel computational method, called PiCNV, which enables to resolve the haplotype sequence composition within CNV regions in nuclear families based on SNP genotyping microarray data. The algorithm allows to i) phase normal and CNV-carrying haplotypes in the copy number variable regions, ii) resolve the allelic copies of rearranged DNA sequence within the haplotypes and iii) infer the heritability of identified haplotypes in trios or larger nuclear families. To our knowledge this is the first program available that can deterministically phase null, mono-, di-, tri- and tetraploid genotypes in CNV loci. We applied our method to study the composition and inheritance of haplotypes in CNV regions of 30 HapMap Yoruban trios and 34 Estonian families. For 93.6% of the CNV loci, PiCNV enabled to unambiguously phase normal and CNV-carrying haplotypes and follow their transmission in the corresponding families. Furthermore, allelic composition analysis identified the co-occurrence of alternative allelic copies within 66.7% of haplotypes carrying copy number gains. We also observed less frequent transmission of CNV-carrying haplotypes from parents to children compared to normal haplotypes and identified an emergence of several de novo deletions and duplications in the offspring.
DNA拷贝数变异（CNV）可改变基因组中特定DNA片段的拷贝数，在人类表型变异和疾病易感性中起着重要作用。已显示许多与基因重叠的CNV会导致多种人类疾病的风险增加，从而凸显了以更高分辨率研究CNV变异性的相关性。到目前为止，还无法确定性地推断CNV区域内存在的不同单倍型的等位基因组成。我们开发了一种名为PiCNV的新型计算方法，该方法能够根据SNP基因分型微阵列数据解析核心家庭中CNV区域内的单倍型序列组成。该算法允许：i）对拷贝数可变区域中的正常和携带CNV的单倍型进行定相；ii）解析单倍型内重排DNA序列的等位基因拷贝；iii）推断三人组或更大核心家庭中已识别单倍型的遗传力。据我们所知，这是第一个可用于确定性地对CNV位点中的无效、单倍体、二倍体、三倍体和四倍体基因型进行定相的程序。我们应用我们的方法研究了30个HapMap约鲁巴三人组和34个爱沙尼亚家庭的CNV区域中单倍型的组成和遗传。对于93.6%的CNV位点，PiCNV能够明确地对正常和携带CNV的单倍型进行定相，并追踪它们在相应家庭中的传递。此外，等位基因组成分析确定了66.7%携带拷贝数增加的单倍型内替代等位基因拷贝的共现。我们还观察到与正常单倍型相比，携带CNV的单倍型从父母向子女的传递频率较低，并在后代中发现了几个新生缺失和重复的出现。
