Genetic and structural variations in the genome account for most of the differences between the genomes of individual humans.

Among these variations, single nucleotide polymorphisms (SNPs) are well characterized and widely used for genome-wide association studies (GWAS) of common human diseases and drug metabolism because of the availability of platforms for large-scale assay, such as microarray systems.

SNPs are divided into several types based on allele frequency in the general population, including rare SNPs (<0.5%) and common SNPs (>1–5%).1 On the basis of the common disease-common variants hypothesis and because of the eraly types of microarrays covering common SNPs, GWAS using common SNPs have been undertaken to identify disease-related alleles.1 However, discoveries from GWAS using common SNPs could explain only 2–15% of heritable variation in disease risk. It has been speculated that there are considerable missing heritability elements.1

Progress in microarray technology and the Human Genome Project resulted in the discovery of variations in gene copy number in the human genome; these variations were termed copy number variations (CNVs).2 CNV is defined as the presence of multiple copies of genome segments >1 kb up to several Mb in size among individuals owing to deletions, insertions, inversions, duplications or complex recombinations.2 More than 7000 CNVs were mapped in the human genome.2 Similar to SNPs, human genome has common and rare CNVs.3

As mapped CNVs were associated with nearly 3000 genes2 and diversity of CNVs was believed to affect human variation, CNVs were expected to account for the missing heritability. Many groups have addressed the presence of CNVs in the human genome and their association with common diseases including neuropsychiatric, infectious, autoimmune, cardiovascular and metabolic diseases.

Recent advances in defining relationships between CNVs and health, diseases and drug response, were summarized by Almal and Padh.3 In their review, the authors clearly and concisely described the general aspects of CNVs and the relationships between human diseases, or drug metabolism and CNVs. To date, many studies have evidenced the involvement of CNVs in many human diseases.

Although mouse models of CNVs were not mentioned in the review, some reports described the establishment of transgenic models of CNVs to validate the effects in mice, which might be important. For example, mouse models of 16p11.2 CNVs exhibited dosage-dependent changes in gene expression, viability, brain architecture and behavior that mimicked autism in humans.4 Therefore, for assessing the effects or associations of CNV, increase in direct experimentation on mice is anticipated.

Till date, a lot of population-based studies have reported that CNVs in some pharmacogenetic genes, such as CYP2D6 and CYP2C19, have a clear role in drug efficacy and toxicity.5 In contrast, some recent GWAS of common CNVs reported rather pessimistic results in common diseases as follows.6, 7

Common CNV-based GWAS in 16 000 cases of eight common diseases—bipolar disorder, breast cancer, coronary artery disease, Crohn's disease, hypertension, rheumatoid arthritis, type-1 diabetes and type-2 diabetes—and 3000 controls revealed that common CNVs were unlikely to contribute to the genetic basis of these diseases.7

In addition, another study based on the assessment of the patterns of linkage disequilibrium between CNVs and SNPs suggested that for complex traits the heritability void left by GWAS could not be compensated by common CNVs.6

As both authors concluded, other platforms or resources of genetic variants could elucidate more disease susceptibilities. Maps of precise nucleotide CNVs and rare CNVs, as well as rare SNPs would be important resources for association studies of human diseases.1

However, the resolution of CNVs on chromosomes (mapping of CNVs) has not been sufficiently high, and has been biased because of limitations in microarray technology. More recently, CNV studies have become more precise to the level of nucleotide resolution and have allowed the discovery of rare CNVs by resequencing personal genomes using the next-generation sequencer.8, 9, 10

The Structural Variation Group of the 1000 Genomes Project constructed a map of CNVs on the basis of whole genome DNA sequencing data from 185 human genomes, which were resequenced using next-generation sequencer.10 The map encompassed 22 035 deletions and 6000 additional structural variations, including insertions and tandem duplications. In addition, more than half of the structural variations were mapped to nucleotide resolution.

Accumulation of such precise and accurate data of CNVs, including personal or rare CNVs, will facilitate disease association studies, and enable the discovery of new disease susceptibilities and drug responses.