Improving power of association tests using multiple sets of imputed genotypes from distributed reference panels.

The accuracy of genotype imputation depends upon two factors: the sample size of the reference panel and the genetic similarity between the reference panel and the target samples. When multiple reference panels are not consented to combine together, it is unclear how to combine the imputation results to optimize the power of genetic association studies. We compared the accuracy of 9,265 Norwegian genomes imputed from three reference panels-1000 Genomes phase 3 (1000G), Haplotype Reference Consortium (HRC), and a reference panel containing 2,201 Norwegian participants from the population-based Nord Trøndelag Health Study (HUNT) from low-pass genome sequencing. We observed that the population-matched reference panel allowed for imputation of more population-specific variants with lower frequency (minor allele frequency (MAF) between 0.05% and 0.5%). The overall imputation accuracy from the population-specific panel was substantially higher than 1000G and was comparable with HRC, despite HRC being 15-fold larger. These results recapitulate the value of population-specific reference panels for genotype imputation. We also evaluated different strategies to utilize multiple sets of imputed genotypes to increase the power of association studies. We observed that testing association for all variants imputed from any panel results in higher power to detect association than the alternative strategy of including only one version of each genetic variant, selected for having the highest imputation quality metric. This was particularly true for lower frequency variants (MAF < 1%), even after adjusting for the additional multiple testing burden.

Lung cancer and DNA repair genes: multilevel association analysis from the International Lung Cancer Consortium.

Lung cancer (LC) is the leading cause of cancer-related death worldwide and tobacco smoking is the major associated risk factor. DNA repair is an important process, maintaining genome integrity and polymorphisms in DNA repair genes may contribute to susceptibility to LC. To explore the role of DNA repair genes in LC, we conducted a multilevel association study with 1655 single nucleotide polymorphisms (SNPs) in 211 DNA repair genes using 6911 individuals pooled from four genome-wide case-control studies. Single SNP association corroborates previous reports of association with rs3131379, located on the gene MSH5 (P = 3.57 × 10-5) and returns a similar risk estimate. The effect of this SNP is modulated by histological subtype. On the log-additive scale, the odds ratio per allele is 1.04 (0.84-1.30) for adenocarcinomas, 1.52 (1.28-1.80) for squamous cell carcinomas and 1.31 (1.09-1.57) for other histologies (heterogeneity test: P = 9.1 × 10(-)(3)). Gene-based association analysis identifies three repair genes associated with LC (P < 0.01): UBE2N, structural maintenance of chromosomes 1L2 and POLB. Two additional genes (RAD52 and POLN) are borderline significant. Pathway-based association analysis identifies five repair pathways associated with LC (P < 0.01): chromatin structure, DNA polymerases, homologous recombination, genes involved in human diseases with sensitivity to DNA-damaging agents and Rad6 pathway and ubiquitination. This first international pooled analysis of a large dataset unravels the role of specific DNA repair pathways in LC and highlights the importance of accounting for gene and pathway effects when studying LC.

Investigation of the fine structure of European populations with applications to disease association studies.

An investigation into fine-scale European population structure was carried out using high-density genetic variation on nearly 6000 individuals originating from across Europe. The individuals were collected as control samples and were genotyped with more than 300 000 SNPs in genome-wide association studies using the Illumina Infinium platform. A major East-West gradient from Russian (Moscow) samples to Spanish samples was identified as the first principal component (PC) of the genetic diversity. The second PC identified a North-South gradient from Norway and Sweden to Romania and Spain. Variation of frequencies at markers in three separate genomic regions, surrounding LCT, HLA and HERC2, were strongly associated with this gradient. The next 18 PCs also accounted for a significant proportion of genetic diversity observed in the sample. We present a method to predict the ethnic origin of samples by comparing the sample genotypes with those from a reference set of samples of known origin. These predictions can be performed using just summary information on the known samples, and individual genotype data are not required. We discuss issues raised by these data and analyses for association studies including the matching of case-only cohorts to appropriate pre-collected control samples for genome-wide association studies.