Identification of large-scale copy number variations in normal population using SNP array Free

39

Since many genes in human cancer cells undergo copy number changes, the dissection of copy number variation in the human genome among normal individuals will be extremely beneficial in order to distinguish the copy number variation (CNV) among normal populations from the abnormal copy number changes of genes related to cancer. We have identified large-scale copy number variations among the normal population from various ethnic groups, based on Affymetrix 50K SNP array data. The high-density SNP probes used in the Affymetrix data set enabled an accurate analysis on large-scale copy number variation, as compared with four previous reports using representational oligonucleotide microarray analysis (ROMA), BAC array CGH, SNP genotyping and fosmid end sequencing. We utilized dChip to infer the copy number of each SNP allele based on the SNP array data and then identified CNVs based on the SNP copy numbers. Our analysis has revealed 155 CNVs, of which about 37% are copy number gains and about 63% are copy number losses among 104 normal individuals. These CNVs are located on more than 100 cytobands across all 23 chromosomes. The CNVs range from 68bp to 18 Mb in length, with a median length of about 86 Kb. An examination of CNVs from each ethnic group has revealed distinct patterns regarding the CNV location and gain/loss ratio among different ethnic groups. In addition, comparison of the CNVs identified by our analysis with those uncovered by previous studies showed that approximately 17% of the 143 unique CNVs discovered by this analysis are concordant with earlier findings by different methods. As part of the program in genomic profiling of pediatric cancers, we take into account all identifiable genetic changes to determine candidate genes and pathways. The results reported here will greatly facilitate the differentiation between copy number variation and somatic changes in tumors. Comparison of the large-scale copy number variations among normal populations to those of cancer patients may lead to a better understanding about the genetic predispositions towards cancer among various ethnic groups. Correlating the large-scale copy number variations with clinical outcomes may also help to refine the ethnically based prognostication of cancers.