Modeling genetic inheritance of copy number variations

Abstract

Copy number variations (CNVs) are being used as genetic markers or functional candidates in gene-mapping studies. However, unlike single nucleotide polymorphism or microsatellite genotyping techniques, most CNV detection methods are limited to detecting total copy numbers, rather than copy number in each of the two homologous chromosomes. To address this issue, we developed a statistical framework for intensity-based CNV detection platforms using family data. Our algorithm identifies CNVs for a family simultaneously, thus avoiding the generation of calls with Mendelian inconsistency while maintaining the ability to detect de novo CNVs. Applications to simulated data and real data indicate that our method significantly improves both call rates and accuracy of boundary inference, compared to existing approaches. We further illustrate the use of Mendelian inheritance to infer SNP allele compositions in each of the two homologous chromosomes in CNV regions using real data. Finally, we applied our method to a set of families genotyped using both the Illumina HumanHap550 and Affymetrix genome-wide 5.0 arrays to demonstrate its performance on both inherited and de novo CNVs. In conclusion, our method produces accurate CNV calls, gives probabilistic estimates of CNV transmission and builds a solid foundation for the development of linkage and association tests utilizing CNVs.

Figure 1.

Illustration of the hidden Markov model framework for modeling genetic inheritance of CNVs in parents–offspring trios. F, M and O represent copy number states of the father, mother and offspring, respectively, and DN is an indicator variable for de novo event status of the offspring.

Figure 2.

Comparative analysis of three CNV calling algorithms on simulated data. For each of the eight scenarios, 1000 trio data sets were simulated and analyzed. We evaluated whether each calling algorithm can identify ‘exactly correct’ CNV calls (calls with the exact CNV boundaries and the exact copy number as true CNVs). The joint-calling algorithm has the overall best performance, especially for inherited duplications.

Figure 3.

Illustration of the signal intensity patterns (LRR and BAF values in upper panel) at a CNV region on 22q11.21 in four members in an AGRE family. This CNV region encompasses the DCGR6 and PRODH gene, as shown in the genome browser (26) shot (lower panel), where the CNV region for each individual is represented by a bar in the browser track (green = three copies, dark green = four copies). With the family information, we can infer that the first child inherits duplications from both the father and the mother, resulting in having four copies of the chromosome region.

Figure 4.

Illustration of a duplication CNV on 10q11.22 that exists in the father and is transmitted to four offspring. The CNV calls are made on six trios separately by the joint-calling algorithm. For each individual, the BAF values for all SNPs within the CNV and the chromosome-specific SNP genotypes (for the first 10 SNPs) are displayed, and the SNP genotypes for the entire region are listed at Supplementary Table 4. The four different parental CNV haplotypes are marked by different colors and denoted by I through IV beneath the genotypes. Combining information from total copy number and the SNP genotypes, we can infer the SNP allele compositions within each homologous chromosome confidently for each offspring.

Figure 5.

Comparison of three CNV calling algorithms in identifying the exact boundaries of the 16p11.2 CNV in offspring of a set of families genotyped by both Illumina HumanHap550 SNP array and Affymetrix genome-wide 5.0 Human SNP array. The CNV calls with exact boundaries were marked by bold font in the table in upper panel. The CNV region is displayed within UCSC genome browser (26), with two tracks representing marker coverage in two different arrays, as well as the RefSeq Genes track showing genes within the CNV. The three Affymetrix CN markers located within segmental duplication regions are marked by a circle and are removed from analysis. Two ∼146 kb flanking segmental duplications are shown as dark orange bars in the Segmental Dups track. The joint-calling algorithm makes more exactly correct CNV calls than the other two calling algorithms.