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

Zhou, Wei; Fritsche, Lars G.; S, Das; Zhang, He; Jb, Nielsen; Holmen, Oddgeir L.; Chen, Jin; Lin, Maoxuan; Elvestad, Maiken B.; Hveem, Kristian; Abecasis, Gonçalo R.; Kang, Hyun Min; Willer, Cristen J.

doi:10.1002/gepi.22067

Cited by 29 publications

(29 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are two strategies for developing reference panels, the first is to use a population with closely matched ancestry to that of the group under study, and the second is to use as many samples as possible. While not evaluated in the context of low-pass sequencing imputation, analysis of DNA arrays shows that reference panels matched to the population of interest outperform diverse reference panels of similar sizes (Mitt et al 2017; Zhou et al 2017; Bai et al 2019; Yoo et al 2019). This would suggest that larger refence panels are preferable as long as they contain sufficient representation of the study population.…”

Section: Discussionmentioning

confidence: 99%

“…These observed effects were for initial population-matched reference panels of ~100 samples, with additional diverse samples increasing the reference panels to over 860 samples (Bai et al 2019). Other analyses compared references panels of > 1500 samples to the Haplotype Reference Consortium (HRC) reference panel (http://www.haplotype-reference-consortium.org/) (McCarthy et al 2016; Mitt et al 2017; Zhou et al 2017; Yoo et al 2019), which consists of 32,611 samples, indicating the potential for increased resolution in human studies compared to canine studies, which used a panel of just 676 samples from 91 breeds (Piras et al 2020). Altogether, at MAFs > 0.05, human imputation studies conducted using DNA array genotypes show non-reference concordance rates > 97.5% and mean r 2 values > 0.95 (Mitt et al 2017; Zhou et al 2017; Yoo et al 2019).…”

Section: Discussionmentioning

confidence: 99%

“…Other analyses compared references panels of > 1500 samples to the Haplotype Reference Consortium (HRC) reference panel (http://www.haplotype-reference-consortium.org/) (McCarthy et al 2016; Mitt et al 2017; Zhou et al 2017; Yoo et al 2019), which consists of 32,611 samples, indicating the potential for increased resolution in human studies compared to canine studies, which used a panel of just 676 samples from 91 breeds (Piras et al 2020). Altogether, at MAFs > 0.05, human imputation studies conducted using DNA array genotypes show non-reference concordance rates > 97.5% and mean r 2 values > 0.95 (Mitt et al 2017; Zhou et al 2017; Yoo et al 2019). By comparison, prior to filtering non-reference concordance rates for low-pass sequence imputation in dogs were at ~ 95% and had mean r 2 values between 0.90 and 0.94 ( Fig 4C ), highlighting the potential gains that can be achieved from improved canine reference panels.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Best practices for analyzing imputed genotypes from low-pass sequencing in dogs

Buckley

Harris

Whitaker

et al. 2021

Preprint

View full text Add to dashboard Cite

Although DNA array-based approaches for genome wide association studies (GWAS) permit the collection of thousands of low-cost genotypes, it is often at the expense of resolution and completeness, as SNP chip technologies are ultimately limited by SNPs chosen during array development. An alternative low-cost approach is low-pass whole genome sequencing (WGS) followed by imputation. Rather than relying on high levels of genotype confidence at a set of select loci, low-pass WGS and imputation relies on the combined information from millions of randomly sampled low confidence genotypes. To investigate low-pass WGS and imputation in the dog, we assessed accuracy and performance by downsampling 97 high-coverage (>15x) WGS datasets from 51 different breeds to approximately 1x coverage, simulating low-pass WGS. Using a reference panel of 676 dogs from 91 breeds, genotypes were imputed from the downsampled data and compared to a truth set of genotypes generated from high coverage WGS. Using our truth set, we optimized a variant quality filtering strategy that retained approximately 80% of 14M imputed sites and lowered the imputation error rate from 3.0% to 1.5%. Seven million sites remained with a MAF > 5% and an average imputation quality score of 0.95. Finally, we simulated the impact of imputation errors on outcomes for case-control GWAS, where small effect sizes were most impacted and medium to large effect sizes were minorly impacted. These analyses provide best practice guidelines for study design and data post-processing of low-pass WGS imputed genotypes in dogs.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Best practices for analyzing imputed genotypes from low-pass sequencing in dogs

Buckley

Harris

Whitaker

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In total, approximately 70,000 HUNT individuals have been genotyped using the Illumina Human CoreExome v1.1 array from both HUNT2 (collected between 1995–1997) and HUNT3 (collected between 2006 and 2008). After quality control of the genotype data, 69,716 European ancestry samples were imputed using a combined Haplotype Reference Consortium reference panel and a population specific whole genome sequence-based imputation panel 25 . 11.2 M single nucleotide variants and 430,000 indels with high imputation quality (imputation R 2 > 0.9) and minor allele count >10 were included in the analysis.…”

Section: Methodsmentioning

confidence: 99%

MEPE loss-of-function variant associates with decreased bone mineral density and increased fracture risk

et al. 2020

Self Cite

View full text Add to dashboard Cite

A major challenge in genetic association studies is that most associated variants fall in the non-coding part of the human genome. We searched for variants associated with bone mineral density (BMD) after enriching the discovery cohort for loss-of-function (LoF) mutations by sequencing a subset of the Nord-Trøndelag Health Study, followed by imputation in the remaining sample (N = 19,705), and identified ten known BMD loci. However, one previously unreported variant, LoF mutation in MEPE, p.(Lys70IlefsTer26, minor allele frequency [MAF] = 0.8%), was associated with decreased ultradistal forearm BMD (P-value = 2.1 × 10−18), and increased osteoporosis (P-value = 4.2 × 10−5) and fracture risk (P-value = 1.6 × 10−5). The MEPE LoF association with BMD and fractures was further evaluated in 279,435 UK (MAF = 0.05%, heel bone estimated BMD P-value = 1.2 × 10−16, any fracture P-value = 0.05) and 375,984 Icelandic samples (MAF = 0.03%, arm BMD P-value = 0.12, forearm fracture P-value = 0.005). Screening for the MEPE LoF mutations before adulthood could potentially prevent osteoporosis and fractures due to the lifelong effect on BMD observed in the study. A key implication for precision medicine is that high-impact functional variants missing from the publicly available cosmopolitan panels could be clinically more relevant than polygenic risk scores.

show abstract

“…For study populations that are not well represented by available reference panels, this problem is exacerbated (Herzig et al, 2018). When imputing genotypes in indigenous and founder populations, or even isolated populations of European ancestry, the results can be significantly improved when a subset of the sample are sequenced and included as a study specific reference panel (SSRP) (Deelen et al, 2014;Hou et al, 2017;Mitt et al, 2017;Pistis et al, 2015;Sidore et al, 2015;Zhou et al, 2017;Herzig et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Kinpute: Using identity by descent to improve genotype imputation

Abney

Sherbiny

2018

Preprint

View full text Add to dashboard Cite

1AbstractMotivationGenotype imputation, though generally accurate, often results in many genotypes being poorly imputed, particularly in studies where the individuals are not well represented by standard reference panels. When individuals in the study share regions of the genome identical by descent (IBD), it is possible to use this information in combination with a study specific reference panel (SSRP) to improve the imputation results. Kinpute uses IBD information—due to either recent, familial relatedness or distant, unknown ancestors— in conjunction with the output from linkage disequilibrium (LD) based imputation methods to compute more accurate genotype probabilities. Kinpute uses a novel method for IBD imputation, which works even in the absence of a pedigree, and results in substantially improved imputation quality.ResultsGiven initial estimates of average IBD between subjects in the study sample, Kinpute uses a novel algorithm to select an optimal set of individuals to sequence and use as an SSRP. Kinpute is designed to use as input both this SSRP and the genotype probabilities output from other LD based imputation software, and uses a new method to combine the LD imputed genotype probabilities with IBD configurations to substantially improve imputation. We tested Kinpute on a human population isolate where 98 individuals have been sequenced. In half of this sample, whose sequence data was masked, we used Impute2 to perform LD based imputation and Kinpute was used to obtain higher accuracy genotype probabilities. Measures of imputation accuracy improved significantly, particularly for those genotypes that Impute2 imputed with low certainty.AvailabilityKinpute is an open-source and freely available C++ software package that can be downloaded from https://github.com/markabney/Kinpute/releases.

show abstract

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

Cited by 29 publications

References 49 publications

Best practices for analyzing imputed genotypes from low-pass sequencing in dogs

Best practices for analyzing imputed genotypes from low-pass sequencing in dogs

MEPE loss-of-function variant associates with decreased bone mineral density and increased fracture risk

Kinpute: Using identity by descent to improve genotype imputation

Contact Info

Product

Resources

About