Comparing genotyping algorithms for Illumina's Infinium whole-genome SNP BeadChips

Background  

Illumina's Infinium SNP BeadChips are extensively used in both small and large-scale genetic studies. A fundamental step in any analysis is the processing of raw allele A and allele B intensities from each SNP into genotype calls (AA, AB, BB). Various algorithms which make use of different statistical models are available for this task. We compare four methods (GenCall, Illuminus, GenoSNP and CRLMM) on data where the true genotypes are known in advance and data from a recently published genome-wide association study.     

An innovative SNP genotyping method adapting to multiple platforms and throughputs

Key message

An innovative genotyping method designated as semi-thermal asymmetric reverse PCR (STARP) was developed for genotyping individual SNPs with improved accuracy, flexible throughputs, low operational costs, and high platform compatibility.

Abstract

Multiplex chip-based technology for genome-scale genotyping of single nucleotide polymorphisms (SNPs) has made great progress in the past two decades. However, PCR-based genotyping of individual SNPs still remains problematic in accuracy, throughput, simplicity, and/or operational costs as well as the compatibility with multiple platforms. Here, we report a novel SNP genotyping method designated semi-thermal asymmetric reverse PCR (STARP). In this method, genotyping assay was performed under unique PCR conditions using two universal priming element-adjustable primers (PEA-primers) and one group of three locus-specific primers: two asymmetrically modified allele-specific primers (AMAS-primers) and their common reverse primer. The two AMAS-primers each were substituted one base in different positions at their 3′ regions to significantly increase the amplification specificity of the two alleles and tailed at 5′ ends to provide priming sites for PEA-primers. The two PEA-primers were developed for common use in all genotyping assays to stringently target the PCR fragments generated by the two AMAS-primers with similar PCR efficiencies and for flexible detection using either gel-free fluorescence signals or gel-based size separation. The state-of-the-art primer design and unique PCR conditions endowed STARP with all the major advantages of high accuracy, flexible throughputs, simple assay design, low operational costs, and platform compatibility. In addition to SNPs, STARP can also be employed in genotyping of indels (insertion–deletion polymorphisms). As vast variations in DNA sequences are being unearthed by many genome sequencing projects and genotyping by sequencing, STARP will have wide applications across all biological organisms in agriculture, medicine, and forensics.

     

Performance comparison of whole-genome sequencing platforms

Whole-genome sequencing is becoming commonplace, but the accuracy and completeness of variant calling by the most widely used platforms from Illumina and Complete Genomics have not been reported. Here we sequenced the genome of an individual with both technologies to a high average coverage of ～76×, and compared their performance with respect to sequence coverage and calling of single-nucleotide variants (SNVs), insertions and deletions (indels). Although 88.1% of the ～3.7 million unique SNVs were concordant between platforms, there were tens of thousands of platform-specific calls located in genes and other genomic regions. In contrast, 26.5% of indels were concordant between platforms. Target enrichment validated 92.7% of the concordant SNVs, whereas validation by genotyping array revealed a sensitivity of 99.3%. The validation experiments also suggested that >60% of the platform-specific variants were indeed present in the genome. Our results have important implications for understanding the accuracy and completeness of the genome sequencing platforms.     

Assessment of two flexible and compatible SNP genotyping platforms: TaqMan SNP Genotyping Assays and the SNPlex Genotyping System

In this review we describe the principles, protocols, and applications of two commercially available SNP genotyping platforms, the TaqMan SNP Genotyping Assays and the SNPlex Genotyping System. Combined, these two technologies meet the requirements of multiple SNP applications in genetics research and pharmacogenetics. We also describe a set of SNP selection tools and validated assay resources which we developed to accelerate the cycle of experimentation on these platforms. Criteria for selecting the more appropriate of these two genotyping technologies are presented: the genetic architecture of the trait of interest, the throughput required, and the number of SNPs and samples needed for a successful study. Overall, the TaqMan assay format is suitable for low- to mid-throughput applications in which a high assay conversion rate, simple assay workflow, and low cost of automation are desirable. The SNPlex Genotyping System, on the other hand, is well suited for SNP applications in which throughput and cost-efficiency are essential, e.g., applications requiring either the testing of large numbers of SNPs and samples, or the flexibility to select various SNP subsets.     

Predicting unobserved phenotypes for complex traits from whole-genome SNP data

Genome-wide association studies (GWAS) for quantitative traits and disease in humans and other species have shown that there are many loci that contribute to the observed resemblance between relatives. GWAS to date have mostly focussed on discovery of genes or regulatory regions habouring causative polymorphisms, using single SNP analyses and setting stringent type-I error rates. Genome-wide marker data can also be used to predict genetic values and therefore predict phenotypes. Here, we propose a Bayesian method that utilises all marker data simultaneously to predict phenotypes. We apply the method to three traits: coat colour, %CD8 cells, and mean cell haemoglobin, measured in a heterogeneous stock mouse population. We find that a model that contains both additive and dominance effects, estimated from genome-wide marker data, is successful in predicting unobserved phenotypes and is significantly better than a prediction based upon the phenotypes of close relatives. Correlations between predicted and actual phenotypes were in the range of 0.4 to 0.9 when half of the number of families was used to estimate effects and the other half for prediction. Posterior probabilities of SNPs being associated with coat colour were high for regions that are known to contain loci for this trait. The prediction of phenotypes using large samples, high-density SNP data, and appropriate statistical methodology is feasible and can be applied in human medicine, forensics, or artificial selection programs.     

Optimizing comparative genomic hybridization probes for genotyping and SNP detection in Plasmodium falciparum

Microarray-based comparative genomic hybridizations (CGH) interrogate genomic DNA to identify structural differences such as amplifications and deletions that are easily detected as large signal aberrations. Subtle signal deviations caused by single nucleotide polymorphisms (SNPs) can also be detected but is challenged by a high AT content (81%) in P. falciparum. We compared genome-wide CGH signal to sequence polymorphisms between parasite strains 3D7, HB3, and Dd2 using NimbleGen microarrays. From 23,191 SNPs (excluding var/rif/stevor genes), our CGH probe set detected SNPs with > 99.9% specificity but low (< 10%) sensitivity. Probe length, melting temperature, GC content, SNP location in the probe, mutation type, and hairpin structures affected SNP sensitivity. Previously unrecognized variable number tandem repeats (VNTRs) also were detected by this method. These findings will guide the redesign of a probe set to optimize an openly available CGH microarray platform for high-resolution genotyping suitable for population genomics studies.     

Inference of kinship coefficients from Korean SNP genotyping data

The determination of relatedness between individuals in a family is crucial in analysis of common complex diseases. We present a method to infer close inter-familial relationships based on SNP genotyping data and provide the relationship coefficient of kinship in Korean families. We obtained blood samples from 43 Korean individuals in two families. SNP data was obtained using the Affymetrix Genome-wide Human SNP array 6.0 and the Illumina Human 1M-Duo chip. To measure the kinship coefficient with the SNP genotyping data, we considered all possible pairs of individuals in each family. The genetic distance between two individuals in a pair was determined using the allele sharing distance method. The results show that genetic distance is proportional to the kinship coefficient and that a close degree of kinship can be confirmed with SNP genotyping data. This study represents the first attempt to identify the genetic distance between very closely related individuals. [BMB Reports 2013; 46(6): 305-309]     

Site frequency spectra from genomic SNP surveys

Genomic survey data now permit an unprecedented level of sensitivity in the detection of departures from canonical evolutionary models, including expansions in population size and selective sweeps. Here, we examine the effects of seemingly subtle differences among sampling distributions on goodness of fit analyses of site frequency spectra constructed from single nucleotide polymorphisms. Conditioning on the observation of exactly two alleles in a random sample results in a site frequency spectrum that is independent of the scaled rate of neutral substitution (θ). Other sampling distributions, including conditioning on a single mutational event in the sample genealogy or randomly selecting a single mutation from a genealogy with multiple mutations, have distinct site frequency spectra that show highly significant departures from the predictions of the biallelic model. Some aspects of data filtering may contribute to significant departures of site frequency spectra from expectation, apart from any violation of the standard neutral model.     

High-throughput genotyping in citrus accessions using an SNP genotyping array

We developed a 384 multiplexed SNP array, named CitSGA-1, for the genotyping of Citrus cultivars, and evaluated the performance and reliability of the genotyping. SNPs were surveyed by direct sequence comparison of the sequence tagged site (STS) fragment amplified from genomic DNA of cultivars representing the genetic diversity of citrus breeding in Japan. Among 1497 SNPs candidates, 384 SNPs for a high-throughput genotyping array were selected based on physical parameters of Illumina’s bead array criteria. The assay using CitSGA-1 was applied to a hybrid population of 88 progeny and 103 citrus accessions for breeding in Japan, which resulted in 73,726 SNP calls. A total of 351 SNPs (91 %) could call different genotypes among the DNA samples, resulting in a success rate for the assay comparable to previously reported rates for other plant species. To confirm the reliability of SNP genotype calls, parentage analysis was applied, and it indicated that the number of reliable SNPs and corresponding STSs were 276 and 213, respectively. The multiplexed SNP genotyping array reported here will be useful for the efficient construction of linkage map, for the detection of markers for marker-assisted breeding, and for the identification of cultivars.     

PCR-based genotyping of SNP markers in sheep

Molecular Biology Reports - Single nucleotide polymorphisms (SNPs) are the main type of variation in genome, enabling them to be associated with traits of economic importance in livestock....     

High-density SNP genotyping detects homogeneity of Spanish and French Basques, and confirms their genomic distinctiveness from other European populations

A recent study reported that Basques do not constitute a genetically distinct population, and that Basques from Spanish and French provinces do not show significant genetic similarity. These conclusions disagree with numerous previous studies, and are not consistent with the historical and linguistic evidence that supports the distinctiveness of Basques. In order to further investigate this controversy, we have genotyped 83 Spanish Basque individuals and used these data to infer population structure based on more than 60,000 single nucleotide polymorphisms of several European populations. Here, we present the first high-throughput analysis including Basques from Spanish and French provinces, and show that all Basques constitute a homogeneous group that can be clearly differentiated from other European populations.     

SNP marker detection and genotyping in tilapia

We have generated a unique resource consisting of nearly 175 000 short contig sequences and 3569 SNP markers from the widely cultured GIFT (Genetically Improved Farmed Tilapia) strain of Nile tilapia (Oreochromis niloticus). In total, 384 SNPs were selected to monitor the wider applicability of the SNPs by genotyping tilapia individuals from different strains and different geographical locations. In all strains and species tested (O. niloticus, O. aureus and O. mossambicus), the genotyping assay was working for a similar number of SNPs (288–305 SNPs). The actual number of polymorphic SNPs was, as expected, highest for individuals from the GIFT population (255 SNPs). In the individuals from an Egyptian strain and in individuals caught in the wild in the basin of the river Volta, 197 and 163 SNPs were polymorphic, respectively. A pairwise calculation of Nei’s genetic distance allowed the discrimination of the individual strains and species based on the genotypes determined with the SNP set. We expect that this set will be widely applicable for use in tilapia aquaculture, e.g. for pedigree reconstruction. In addition, this set is currently used for assaying the genetic diversity of native Nile tilapia in areas where tilapia is, or will be, introduced in aquaculture projects. This allows the tracing of escapees from aquaculture and the monitoring of effects of introgression and hybridization.     

Automated SNP genotype clustering algorithm to improve data completeness in high-throughput SNP genotyping datasets from custom arrays

High-throughput SNP genotyping platforms use automated genotype calling algorithms to assign genotypes. While these algorithms work efficiently for individual platforms, they are not compatible with other platforms, and have individual biases that result in missed genotype calls. Here we present data on the use of a second complementary SNP genotype clustering algorithm. The algorithm was originally designed for individual fluorescent SNP genotyping assays, and has been optimized to permit the clustering of large datasets generated from custom-designed Affymetrix SNP panels. In an analysis of data from a 3K array genotyped on 1,560 samples, the additional analysis increased the overall number of genotypes by over 45,000, significantly improving the completeness of the experimental data. This analysis suggests that the use of multiple genotype calling algorithms may be advisable in high-throughput SNP genotyping experiments. The software is written in Perl and is available from the corresponding author.     

单核苷酸多态性基因分型技术原理与进展

在基因组规模了解遗传变异与生物功能之间的关系可望为生物学带来全新的深入认识。本从等位基因分型机理、反应形式和检测方法等三个方面讨论SNP分型方法的现状,并简要介绍了目前应用的一些分型方法。     

Large SNP arrays for genotyping in crop plants

Genotyping with large numbers of molecular markers is now an indispensable tool within plant genetics and breeding. Especially through the identification of large numbers of single nucleotide polymorphism (SNP) markers using the novel high-throughput sequencing technologies, it is now possible to reliably identify many thousands of SNPs at many different loci in a given plant genome. For a number of important crop plants, SNP markers are now being used to design genotyping arrays containing thousands of markers spread over the entire genome and to analyse large numbers of samples. In this article, we discuss aspects that should be considered during the design of such large genotyping arrays and the analysis of individuals. The fact that crop plants are also often autopolyploid or allopolyploid is given due consideration. Furthermore, we outline some potential applications of large genotyping arrays including high-density genetic mapping, characterization (fingerprinting) of genetic material and breeding-related aspects such as association studies and genomic selection.     

Calibrating the performance of SNP arrays for whole-genome association studies

To facilitate whole-genome association studies (WGAS), several high-density SNP genotyping arrays have been developed. Genetic coverage and statistical power are the primary benchmark metrics in evaluating the performance of SNP arrays. Ideally, such evaluations would be done on a SNP set and a cohort of individuals that are both independently sampled from the original SNPs and individuals used in developing the arrays. Without utilization of an independent test set, previous estimates of genetic coverage and statistical power may be subject to an overfitting bias. Additionally, the SNP arrays' statistical power in WGAS has not been systematically assessed on real traits. One robust setting for doing so is to evaluate statistical power on thousands of traits measured from a single set of individuals. In this study, 359 newly sampled Americans of European descent were genotyped using both Affymetrix 500K (Affx500K) and Illumina 650Y (Ilmn650K) SNP arrays. From these data, we were able to obtain estimates of genetic coverage, which are robust to overfitting, by constructing an independent test set from among these genotypes and individuals. Furthermore, we collected liver tissue RNA from the participants and profiled these samples on a comprehensive gene expression microarray. The RNA levels were used as a large-scale set of quantitative traits to calibrate the relative statistical power of the commercial arrays. Our genetic coverage estimates are lower than previous reports, providing evidence that previous estimates may be inflated due to overfitting. The Ilmn650K platform showed reasonable power (50% or greater) to detect SNPs associated with quantitative traits when the signal-to-noise ratio (SNR) is greater than or equal to 0.5 and the causal SNP's minor allele frequency (MAF) is greater than or equal to 20% (N=359). In testing each of the more than 40,000 gene expression traits for association to each of the SNPs on the Ilmn650K and Affx500K arrays, we found that the Ilmn650K yielded 15% times more discoveries than the Affx500K at the same false discovery rate (FDR) level.     

Comparative whole-genome hybridization reveals genomic islands in Brucella species

Brucella species are responsible for brucellosis, a worldwide zoonotic disease causing abortion in domestic animals and Malta fever in humans. Based on host preference, the genus is divided into six species. Brucella abortus, B. melitensis, and B. suis are pathogenic to humans, whereas B. ovis and B. neotomae are nonpathogenic to humans and B. canis human infections are rare. Limited genome diversity exists among Brucella species. Comparison of Brucella species whole genomes is, therefore, likely to identify factors responsible for differences in host preference and virulence restriction. To facilitate such studies, we used the complete genome sequence of B. melitensis 16M, the species highly pathogenic to humans, to construct a genomic microarray. Hybridization of labeled genomic DNA from Brucella species to this microarray revealed a total of 217 open reading frames (ORFs) altered in five Brucella species analyzed. These ORFs are often found in clusters (islands) in the 16M genome. Examination of the genomic context of these islands suggests that many are horizontally acquired. Deletions of genetic content identified in Brucella species are conserved in multiple strains of the same species, and genomic islands missing in a given species are often restricted to that particular species. These findings suggest that, whereas the loss or gain of genetic material may be related to the host range and virulence restriction of certain Brucella species for humans, independent mechanisms involving gene inactivation or altered expression of virulence determinants may also contribute to these differences.     

High-throughput SNP genotyping with the GoldenGate assay in maize

Single nucleotide polymorphisms (SNPs) are abundant and evenly distributed throughout the genomes of most plant species. They have become an ideal marker system for genetic research in many crops. Several high throughput platforms have been developed that allow rapid and simultaneous genotyping of up to a million SNP markers. In this study, a custom GoldenGate assay containing 1,536 SNPs was developed based on public SNP information for maize and used to genotype two recombinant inbred line (RIL) populations (Zong3 x 87-1, and B73 x By804) and a panel of 154 diverse inbred lines. Over 90% of the SNPs were successfully scored in the diversity panel and the two RIL populations, with a genotyping error rate of less than 2%. A total of 975 SNP markers detected polymorphism in at least one of the two mapping populations, with a polymorphic rate of 38.5% in Zong3 x 87-1 and 52.6% in B73 x By804. The polymorphic SNPs in B73 x By804 have been integrated with previously mapped simple sequence repeat markers to construct a high-density linkage map containing 662 markers with a total length of 1,673.7 cM and an average of 2.53 cM between two markers. The minor allelic frequency (MAF) was distributed evenly across 10 continued classes from 0.05 to 0.5, and about 16% of the SNP markers had a MAF below 10% in the diversity panel. Polymorphism rates for individual SNP markers in pair-wise comparisons of genotypes tested ranged from 0.3 to 63.8% with an average of 36.3%. Most SNPs used in this GoldenGate assay appear to be equally useful for diversity analysis, marker-trait association studies, and marker-aided breeding.