Accurate Estimates Of Allele Frequencies Research Articles

High-throughput estimation of allele frequencies using combined pooled-population sequencing and haplotype-based data processing

BackgroundIn addition to heterogeneity and artificial selection, natural selection is one of the forces used to combat climate change and improve agrobiodiversity in evolutionary plant breeding. Accurate identification of the specific genomic effects of natural selection will likely accelerate transfer between populations. Thus, insights into changes in allele frequency, adequate population size, gene flow and drift are essential. However, observing such effects often involves a trade-off between costs and resolution when a large sample of genotypes for many loci is analysed. Pool genotyping approaches achieve high resolution and precision in estimating allele frequency when sequence coverage is high. Nevertheless, high-coverage pool sequencing of large genomes is expensive.ResultsThree pool samples (n = 300, 300, 288) from a barley backcross population were generated to assess the population's allele frequency. The tested population (BC2F21) has undergone 18 generations of natural adaption to conventional farming practice. The accuracies of estimated pool-based allele frequencies and genome coverage yields were compared using three next-generation sequencing genotyping methods. To achieve accurate allele frequency estimates with low sequence coverage, we employed a haplotyping approach. Low coverage allele frequencies of closely located single polymorphisms were aggregated into a single haplotype allele frequency, yielding 2-to-271-times higher depth and increased precision. When we combined different haplotyping tactics, we found that gene and chip marker-based haplotype analyses performed equivalently or better compared with simple contig haplotype windows. Comparing multiple pool samples and referencing against an individual sequencing approach revealed that whole-genome pool re-sequencing (WGS) achieved the highest correlation with individual genotyping (≥ 0.97). In contrast, transcriptome-based genotyping (MACE) and genotyping by sequencing (GBS) pool replicates were significantly associated with higher error rates and lower correlations, but are still valuable to detect large allele frequency variations.ConclusionsThe proposed strategy identified the allele frequency of populations with high accuracy at low cost. This is particularly relevant to evolutionary plant breeding of crops with very large genomes, such as barley. Whole-genome low coverage re-sequencing at 0.03 × coverage per genotype accurately estimated the allele frequency when a loci-based haplotyping approach was applied. The implementation of annotated haplotypes capitalises on the biological background and statistical robustness.

Accurate Allele Frequencies from Ultra-low Coverage Pool-Seq Samples in Evolve-and-Resequence Experiments.

Evolve-and-resequence (E+R) experiments leverage next-generation sequencing technology to track the allele frequency dynamics of populations as they evolve. While previous work has shown that adaptive alleles can be detected by comparing frequency trajectories from many replicate populations, this power comes at the expense of high-coverage (>100x) sequencing of many pooled samples, which can be cost-prohibitive. Here, we show that accurate estimates of allele frequencies can be achieved with very shallow sequencing depths (<5x) via inference of known founder haplotypes in small genomic windows. This technique can be used to efficiently estimate frequencies for any number of bi-allelic SNPs in populations of any model organism founded with sequenced homozygous strains. Using both experimentally-pooled and simulated samples of Drosophila melanogaster, we show that haplotype inference can improve allele frequency accuracy by orders of magnitude for up to 50 generations of recombination, and is robust to moderate levels of missing data, as well as different selection regimes. Finally, we show that a simple linear model generated from these simulations can predict the accuracy of haplotype-derived allele frequencies in other model organisms and experimental designs. To make these results broadly accessible for use in E+R experiments, we introduce HAF-pipe, an open-source software tool for calculating haplotype-derived allele frequencies from raw sequencing data. Ultimately, by reducing sequencing costs without sacrificing accuracy, our method facilitates E+R designs with higher replication and resolution, and thereby, increased power to detect adaptive alleles.

Abstract 3280: Utah somatic variant calling pipeline featuring multi-sample joint calling, variant-graph based accurate allele frequency estimation and subclone analysis

Abstract Current tumor variant detection software tools are focused on the identification of somatic mutations in a single tumor sample (e.g. the primary tumor) compared to normal control tissue, and are not adequate for studies where multiple samples from the same patient, either at consecutive time points or at different metastatic sites are collected. Here we present a software framework that facilitates simultaneous analysis of such multi-sample datasets. Our framework utilizes our FreeBayes variant caller for jointly calling all tumor samples and normal controls from the patient. This ensures that even very low frequency mutations are called because read evidence is aggregated across all samples; and that allele frequency information for a called somatic variant is provided uniformly for all samples. Short and medium-length INDELs, and structural variants are identified by our RUFUS software, a new reference-free mutation calling algorithm for both germline and somatic mutation detection, with extremely high sensitivity and specificity. We use our Graphite tool, a graph-genome based read realignment algorithm, to eliminate false positive somatic calls i.e. inherited variants masquerading as somatic mutations, and to substantially increase accuracy of allele frequency estimation of the true somatic calls, a feature critical for downstream subclone analysis. In addition, we use the FACETS package for CNV and LOH analysis, also critical for downstream subclone analysis. Finally, we utilize SubcloneSeeker, our algorithm to reconstruct the evolution of the cancer at the subclonal level from the somatic variant allele frequencies, to understand how the tumor evolved over time, in response to multiple courses of treatment, or over space, across multiple metastatic sites. We demonstrate the use of this pipeline in published and ongoing studies involving longitudinal and multi-site tumor sample datasets. Citation Format: Yi Qiao, Xiaomeng Huang, Dillon Lee, Andrew Farrell, Thomas Nicholas, Brent Pederson, Aaron Quinlan, Gabor Marth. Utah somatic variant calling pipeline featuring multi-sample joint calling, variant-graph based accurate allele frequency estimation and subclone analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3280.

Clear: Composition of Likelihoods for Evolve and Resequence Experiments.

The advent of next generation sequencing technologies has made whole-genome and whole-population sampling possible, even for eukaryotes with large genomes. With this development, experimental evolution studies can be designed to observe molecular evolution “in action” via evolve-and-resequence (E&R) experiments. Among other applications, E&R studies can be used to locate the genes and variants responsible for genetic adaptation. Most existing literature on time-series data analysis often assumes large population size, accurate allele frequency estimates, or wide time spans. These assumptions do not hold in many E&R studies. In this article, we propose a method—composition of likelihoods for evolve-and-resequence experiments (Clear)—to identify signatures of selection in small population E&R experiments. Clear takes whole-genome sequences of pools of individuals as input, and properly addresses heterogeneous ascertainment bias resulting from uneven coverage. Clear also provides unbiased estimates of model parameters, including population size, selection strength, and dominance, while being computationally efficient. Extensive simulations show that Clear achieves higher power in detecting and localizing selection over a wide range of parameters, and is robust to variation of coverage. We applied the Clear statistic to multiple E&R experiments, including data from a study of adaptation of Drosophila melanogaster to alternating temperatures and a study of outcrossing yeast populations, and identified multiple regions under selection with genome-wide significance.

VarDict: a novel and versatile variant caller for next-generation sequencing in cancer research.

Accurate variant calling in next generation sequencing (NGS) is critical to understand cancer genomes better. Here we present VarDict, a novel and versatile variant caller for both DNA- and RNA-sequencing data. VarDict simultaneously calls SNV, MNV, InDels, complex and structural variants, expanding the detected genetic driver landscape of tumors. It performs local realignments on the fly for more accurate allele frequency estimation. VarDict performance scales linearly to sequencing depth, enabling ultra-deep sequencing used to explore tumor evolution or detect tumor DNA circulating in blood. In addition, VarDict performs amplicon aware variant calling for polymerase chain reaction (PCR)-based targeted sequencing often used in diagnostic settings, and is able to detect PCR artifacts. Finally, VarDict also detects differences in somatic and loss of heterozygosity variants between paired samples. VarDict reprocessing of The Cancer Genome Atlas (TCGA) Lung Adenocarcinoma dataset called known driver mutations in KRAS, EGFR, BRAF, PIK3CA and MET in 16% more patients than previously published variant calls. We believe VarDict will greatly facilitate application of NGS in clinical cancer research.

Abstract 4864: VarDict: A novel and versatile variant caller for next-generation sequencing in cancer research

Abstract Cancer genomes are known to harbor a wide range of mutations, including complex variants with combination of insertion and deletion (InDels). Next generation sequencing (NGS) has revolutionized our understanding of mutations in cancer. Most of current variant callers from NGS focused on single nucleotide variants (SNV) or short InDels. However, detection of complex variants remains a challenge and off limits to most current variant callers. In addition, efficient analysis of ultra-deep targeted sequencing without downsampling for low frequency mutations in heterogeneous cancer samples, which is becoming more routine, is also a challenge and not handled well by most current variant callers. Here, we describe VarDict, a novel and versatile variant caller for ultra-deep targeted deep sequencing, exome, whole genome, and RNA-seq. VarDict is designed for heterogeneous cancer genomes and is able to simultaneously call SNVs (Single-Nucleotide Variants), MNVs (Multiple-Nucleotide Variants), InDels (user-defined sizes), and complex variants (combination of aforementioned events). VarDict handles ultra-deep sequencing of runs up to mean coverage of 1M without down-sampling or significant loss of performance. It performs local realignment around InDels on the fly and rescues soft-clipped reads for more accurate estimation of allele frequencies as well as allowing calls of InDels only supported by soft-clipped reads. In addition, it performs amplicon aware variant calling for PCR-based targeted sequencing by avoiding calling variants in PCR primers, discounting primer depths, and importantly, detecting variants with amplicon-bias, a common artifact for PCR based targeted sequencing. VarDict can also be run in paired mode to identify somatic or LOH variants, as well as variants whose allele frequencies have shifted significantly. It thus enables paired DNA-seq and RNA-seq variant calling that most variant callers do not handle well. To demonstrate the value of VarDict in practice, we applied VarDict on the WGS of NA12878, and compared the result to the calls made in Genome In A Bottle (GiaB). VarDict was able to call &amp;gt;96% of variants in GiaB. In addition, it found many more variants likely missed by GiaB, especially complex variants, many of which were never categorized before. We further applied VarDict in the ICGC-TCGA DREAM Mutation Calling challenge (syn312572). We found it to be as sensitive as the more commonly used somatic SNP callers like MuTect, Freebayes, and VarScan in calling SNPs but more sensitive in calling InDels than an array of other variant callers, including VarScan, FreeBayes, and Scalpel. VarDict is fully open source, implemented in Perl, and uses memory efficiently, regardless of the depth, making it a HPC cluster friendly tool. VarDict has further been integrated into bcbio-nextgen, an open source framework for scalable NGS analysis for ease of deployment. VarDict is freely available in GitHub (https://github.com/AstraZeneca-NGS/VarDict). Citation Format: Zhongwu Lai, Aleksandra Markovets, Miika Ahdesmaki, Justin Johnson. VarDict: A novel and versatile variant caller for next-generation sequencing in cancer research. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4864. doi:10.1158/1538-7445.AM2015-4864

Abstract 3439: Recurrent CDC25C mutations drive malignant transformation in FPD/AML

Abstract Germline mutations in RUNX1 cause familial platelet disorder with predisposit ion to acute myelogenous leukemia (FPD/AML) which is characterized by platelet defects and hematological malignancies. The molecular pathogenesis is poorly understood because of the rarity of this disease, which has impeded the clinical improvement to refine cancer therapy and to identify patients at risk for the onset of malignancies in FPD/AML. Thus, we conducted a nationwide survey to identify patients with familial thrombocytopenia or hematological malignancies. After screening germline RUNX1 mutations in 73 index patients, 13 patients in 7 pedigrees were diagnosed as having FPD/AML, of which 7 had developed hematological malignancies. The mutational spectrum of FPD/AML was unveiled by whole-exome sequencing and targeted deep sequencing, and we identified CDC25C mutation as a recurrent target of FPD/AML, which was found in 7 of 13 (53 %) patients, including 4 FPD/AML patients who had not developed malignancies, although the variant allele frequencies were much lower than those found in patients with malignancies. Furthermore, GATA2 mutations were subclonal in 3 of 7 individuals with CDC25C abnormalities. Deep sequencing enabled an accurate estimation of allele frequencies for individual mutations and prediction of hierarchical architecture, which disclosed that CDC25C mutations form a founding pre-leukemic clone, followed by stepwise acquisition of subclonal mutations that contributed to leukemia progression. Single cell sequencing validated the estimation of the clonal evolution model. In all of the mutated forms of CDC25C, their binding capacity with c-TAK1 was reduced, resulting in decreased phosphorylation of CDC25C at Ser216. As a consequence, mutated CDC25C failed to bind 14-3-3 protein efficiently and stayed in the nucleus at G1/S phase. In accordance with these observations, CDC25C mutants enhanced mitosis entry, which was exaggerated by low dose radiation-induced DNA damage. These results suggest that CDC25C mutation results in disruption of the DNA checkpoint machinery, indicating that CDC25C is a novel target in human cancer. It is known that FPD/AML-associated RUNX1 mutations evokes DNA damage and activates the DNA checkpoint mechanism. We found, however, that introduction of CDC25C mutation results in the marked enhancement of mitotic entry in spite of co-expression of RUNX1 mutation in Ba/F3 cells. Thus, premature mitosis by loss of DNA checkpoint mechanisms in the presence of mutated CDC25C may contribute to malignant transformation of RUNX1-mutated cells. In conclusion, malignant transformation of FPD/AML is formed by stepwise acquisition of mutations and clonal selection, which is initiated by a CDC25C mutation in the pre-leukemic phase and further driven by mutations in GATA2 or other genes. These findings will facilitate the diagnosis and monitoring of patients with FPD/AML who are at an increased risk of developing hematological malignancy. Citation Format: Akihide Yoshimi, Takashi Toya, Masahito Kawazu, Toshihide Ueno, Ayato Tsukamoto, Hiromitsu Iizuka, Masahiro Nakagawa, Yasuhito Nannya, Shunya Arai, Motoshi Ichikawa, Hironori Harada, Kensuke Usuki, Yasuhide Hayashi, Etsuro Ito, Keita Kirito, Hideaki Nakajima, Hiroyuki Mano, Mineo Kurokawa. Recurrent CDC25C mutations drive malignant transformation in FPD/AML. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3439. doi:10.1158/1538-7445.AM2014-3439

Assessing the effect of sequencing depth and sample size in population genetics inferences.

Next-Generation Sequencing (NGS) technologies have dramatically revolutionised research in many fields of genetics. The ability to sequence many individuals from one or multiple populations at a genomic scale has greatly enhanced population genetics studies and made it a data-driven discipline. Recently, researchers have proposed statistical modelling to address genotyping uncertainty associated with NGS data. However, an ongoing debate is whether it is more beneficial to increase the number of sequenced individuals or the per-sample sequencing depth for estimating genetic variation. Through extensive simulations, I assessed the accuracy of estimating nucleotide diversity, detecting polymorphic sites, and predicting population structure under different experimental scenarios. Results show that the greatest accuracy for estimating population genetics parameters is achieved by employing a large sample size, despite single individuals being sequenced at low depth. Under some circumstances, the minimum sequencing depth for obtaining accurate estimates of allele frequencies and to identify polymorphic sites is , where both alleles are more likely to have been sequenced. On the other hand, inferences of population structure are more accurate at very large sample sizes, even with extremely low sequencing depth. This all points to the conclusion that under various experimental scenarios, in cost-limited population genetics studies, large sample sizes at low sequencing depth are desirable to achieve high accuracy. These findings will help researchers design their experimental set-ups and guide further investigation on the effect of protocol design for genetic research.

Cost-effective genome-wide estimation of allele frequencies from pooled DNA in Atlantic salmon (Salmo salarL.)

BackgroundNew sequencing technologies have tremendously increased the number of known molecular markers (single nucleotide polymorphisms; SNPs) in a variety of species. Concurrently, improvements to genotyping technology have now made it possible to efficiently genotype large numbers of genome-wide distributed SNPs enabling genome wide association studies (GWAS). However, genotyping significant numbers of individuals with large number of SNPs remains prohibitively expensive for many research groups. A possible solution to this problem is to determine allele frequencies from pooled DNA samples, such ‘allelotyping’ has been presented as a cost-effective alternative to individual genotyping and has become popular in human GWAS. In this article we have tested the effectiveness of DNA pooling to obtain accurate allele frequency estimates for Atlantic salmon (Salmo salar L.) populations using an Illumina SNP-chip.ResultsIn total, 56 Atlantic salmon DNA pools from 14 populations were analyzed on an Atlantic salmon SNP-chip containing probes for 5568 SNP markers, 3928 of which were bi-allelic. We developed an efficient quality control filter which enables exclusion of loci showing high error rate and minor allele frequency (MAF) close to zero. After applying multiple quality control filters we obtained allele frequency estimates for 3631 bi-allelic loci. We observed high concordance (r > 0.99) between allele frequency estimates derived from individual genotyping and DNA pools. Our results also indicate that even relatively small DNA pools (35 individuals) can provide accurate allele frequency estimates for a given sample.ConclusionsDespite of higher level of variation associated with array replicates compared to pool construction, we suggest that both sources of variation should be taken into account. This study demonstrates that DNA pooling allows fast and high-throughput determination of allele frequencies in Atlantic salmon enabling cost-efficient identification of informative markers for discrimination of populations at various geographical scales, as well as identification of loci controlling ecologically and economically important traits.

Reducing bias of allele frequency estimates by modeling SNP genotype data with informative missingness.

The presence of missing single-nucleotide polymorphism (SNP) genotypes is common in genetic studies. For studies with low-density SNPs, the most commonly used approach to dealing with genotype missingness is to simply remove the observations with missing genotypes from the analyses. This naïve method is straightforward but is valid only when the missingness is random. However, a given assay often has a different capability in genotyping heterozygotes and homozygotes, causing the phenomenon of “differential dropout” in the sense that the missing rates of heterozygotes and homozygotes are different. In practice, differential dropout among genotypes exists in even carefully designed studies, such as the data from the HapMap project and the Wellcome Trust Case Control Consortium. Under the assumption of Hardy–Weinberg equilibrium and no genotyping error, we here propose a statistical method to model the differential dropout among different genotypes. Compared with the naïve method, our method provides more accurate allele frequency estimates when the differential dropout is present. To demonstrate its practical use, we further apply our method to the HapMap data and a scleroderma data set.

The Next Generation of Molecular Markers From Massively Parallel Sequencing of Pooled DNA Samples

Next generation sequencing (NGS) is about to revolutionize genetic analysis. Currently NGS techniques are mainly used to sequence individual genomes. Due to the high sequence coverage required, the costs for population-scale analyses are still too high to allow an extension to nonmodel organisms. Here, we show that NGS of pools of individuals is often more effective in SNP discovery and provides more accurate allele frequency estimates, even when taking sequencing errors into account. We modify the population genetic estimators Tajima's π and Watterson's to obtain unbiased estimates from NGS pooling data. Given the same sequencing effort, the resulting estimators often show a better performance than those obtained from individual sequencing. Although our analysis also shows that NGS of pools of individuals will not be preferable under all circumstances, it provides a cost-effective approach to estimate allele frequencies on a genome-wide scale.

Resequencing of pooled DNA for detecting disease associations with rare variants

A combination of common and rare variants is thought to contribute to genetic susceptibility to complex diseases. Recently, next-generation sequencers have greatly lowered sequencing costs, providing an opportunity to identify rare disease variants in large genetic epidemiology studies. At present, it is still expensive and time consuming to resequence large number of individual genomes. However, given that next-generation sequencing technology can provide accurate estimates of allele frequencies from pooled DNA samples, it is possible to detect associations of rare variants using pooled DNA sequencing. Current statistical approaches to the analysis of associations with rare variants are not designed for use with pooled next-generation sequencing data. Hence, they may not be optimal in terms of both validity and power. Therefore, we propose here a new statistical procedure to analyze the output of pooled sequencing data. The test statistic can be computed rapidly, making it feasible to test the association of a large number of variants with disease. By simulation, we compare this approach to Fisher's exact test based either on pooled or individual genotypic data. Our results demonstrate that the proposed method provides good control of the Type I error rate, while yielding substantially higher power than Fisher's exact test using pooled genotypic data for testing rare variants, and has similar or higher power than that of Fisher's exact test using individual genotypic data. Our results also provide guidelines on how various parameters of the pooled sequencing design affect the efficiency of detecting associations.

A new analysis tool for individual-level allele frequency for genomic studies

BackgroundAllele frequency is one of the most important population indices and has been broadly applied to genetic/genomic studies. Estimation of allele frequency using genotypes is convenient but may lose data information and be sensitive to genotyping errors.ResultsThis study utilizes a unified intensity-measuring approach to estimating individual-level allele frequencies for 1,104 and 1,270 samples genotyped with the single-nucleotide-polymorphism arrays of the Affymetrix Human Mapping 100K and 500K Sets, respectively. Allele frequencies of all samples are estimated and adjusted by coefficients of preferential amplification/hybridization (CPA), and large ethnicity-specific and cross-ethnicity databases of CPA and allele frequency are established. The results show that using the CPA significantly improves the accuracy of allele frequency estimates; moreover, this paramount factor is insensitive to the time of data acquisition, effect of laboratory site, type of gene chip, and phenotypic status. Based on accurate allele frequency estimates, analytic methods based on individual-level allele frequencies are developed and successfully applied to discover genomic patterns of allele frequencies, detect chromosomal abnormalities, classify sample groups, identify outlier samples, and estimate the purity of tumor samples. The methods are packaged into a new analysis tool, ALOHA (Allele-frequency/Loss-of-heterozygosity/Allele-imbalance).ConclusionsThis is the first time that these important genetic/genomic applications have been simultaneously conducted by the analyses of individual-level allele frequencies estimated by a unified intensity-measuring approach. We expect that additional practical applications for allele frequency analysis will be found. The developed databases and tools provide useful resources for human genome analysis via high-throughput single-nucleotide-polymorphism arrays. The ALOHA software was written in R and R GUI and can be downloaded at http://www.stat.sinica.edu.tw/hsinchou/genetics/aloha/ALOHA.htm.

Efficient Methods to Compute Genomic Predictions

Efficient methods for processing genomic data were developed to increase reliability of estimated breeding values and to estimate thousands of marker effects simultaneously. Algorithms were derived and computer programs tested with simulated data for 2,967 bulls and 50,000 markers distributed randomly across 30 chromosomes. Estimation of genomic inbreeding coefficients required accurate estimates of allele frequencies in the base population. Linear model predictions of breeding values were computed by 3 equivalent methods: 1) iteration for individual allele effects followed by summation across loci to obtain estimated breeding values, 2) selection index including a genomic relationship matrix, and 3) mixed model equations including the inverse of genomic relationships. A blend of first- and second-order Jacobi iteration using 2 separate relaxation factors converged well for allele frequencies and effects. Reliability of predicted net merit for young bulls was 63% compared with 32% using the traditional relationship matrix. Nonlinear predictions were also computed using iteration on data and nonlinear regression on marker deviations; an additional (about 3%) gain in reliability for young bulls increased average reliability to 66%. Computing times increased linearly with number of genotypes. Estimation of allele frequencies required 2 processor days, and genomic predictions required <1 d per trait, and traits were processed in parallel. Information from genotyping was equivalent to about 20 daughters with phenotypic records. Actual gains may differ because the simulation did not account for linkage disequilibrium in the base population or selection in subsequent generations.

No evidence for multiple loci affecting rheumatoid arthritis risk on chromosome 6p21

The influence of certain alleles of the HLA-DRB1 locus on risk for rheumatoid arthritis has been well established through linkage and association studies. In addition, other loci in the HLA region on 6p21 may also affect an individual's risk profile. Here, we used a method to detect excess identity-by-descent sharing between affected sib pairs conditional on the observed genotypes at the hypothesized causal locus to test for the presence of additional arthritis risk loci in the linked region. We used affected sib pairs from two different studies. Because the test depends heavily on specifying accurate allele frequency estimates at the proposed causal locus, we used HLA-DRB1 allele frequency estimates from a large, population-based sample. We also discuss an alternate form of the test in which we could condition on parental genotypes, thereby eliminating the need for actual allele frequencies. The test showed no evidence for the presence of additional arthritis risk loci in the region in the British or North American samples made available for Genetic Analysis Workshop 15. Given the prior knowledge that there likely are arthritis risk loci other than HLA-DRB1 in the region, it appears the tests may have inadequate power to detect the presence of these loci in certain cases.

PPC: an algorithm for accurate estimation of SNP allele frequencies in small equimolar pools of DNA using data from high density microarrays

Robust estimation of allele frequencies in pools of DNA has the potential to reduce genotyping costs and/or increase the number of individuals contributing to a study where hundreds of thousands of genetic markers need to be genotyped in very large populations sample sets, such as genome wide association studies. In order to make accurate allele frequency estimations from pooled samples a correction for unequal allele representation must be applied. We have developed the polynomial based probe specific correction (PPC) which is a novel correction algorithm for accurate estimation of allele frequencies in data from high-density microarrays. This algorithm was validated through comparison of allele frequencies from a set of 10 individually genotyped DNA's and frequencies estimated from pools of these 10 DNAs using GeneChip 10K Mapping Xba 131 arrays. Our results demonstrate that when using the PPC to correct for allelic biases the accuracy of the allele frequency estimates increases dramatically.

Streamlined analysis of pooled genotype data in SNP‐based association studies

Several groups have developed methods for estimating allele frequencies in DNA pools as a fast and cheap way for detecting allelic association between genetic markers and disease. To obtain accurate estimates of allele frequencies, a correction factor k for the degree to which measurement of allele-specific products is biased is generally applied. Factor k is usually obtained as the ratio of the two allele-specific signals in samples from heterozygous individuals, a step that can significantly impair throughput and increase cost. We have systematically investigated the properties of k through the use of empirical and simulated data. We show that for the dye terminator primer extension genotyping method we have applied, the correction factor k is substantially influenced by the dye terminators incorporated, but also by the terminal 3' base of the extension primer. We also show that the variation in k is large enough to result in unacceptable error rates if association studies are conducted without regard to k. We show that the impact of ignoring k can be neutralized by applying a correction factor k(max) that can be easily derived, but this at the potential cost of an increase in type I error. Finally, based upon observed distributions for k we derive a method allowing the estimation of the probability pooled data reflects significant differences in the allele frequencies between the subjects comprising the pools. By controlling the error rates in the absence of knowledge of the appropriate SNP-specific correction factors, each approach enhances the performance of DNA pooling, while considerably streamlining the method by reducing time and cost.

A central resource for accurate allele frequency estimation from pooled DNA genotyped on DNA microarrays

Analysing pooled DNA on microarrays is an efficient way to genotype hundreds of individuals for thousands of markers for genome-wide association. Although direct comparison of case and control fluorescence scores is possible, correction for differential hybridization of alleles is important, particularly for rare single nucleotide polymorphisms. Such correction relies on heterozygous fluorescence scores and requires the genotyping of hundreds of individuals to obtain sufficient estimates of the correction factor, completely negating any benefit gained by pooling samples. We explore the effect of differential hybridization on test statistics and provide a solution to this problem in the form of a central resource for the accumulation of heterozygous fluorescence scores, allowing accurate allele frequency estimation at no extra cost.

RAPD and AFLP assessment of genetic variation in a landrace of pepper (Capsicum annuum L.), grown in North-West Italy

In several regions of Italy as well as other parts of southern Europe, the heterogeneity of the land, the climate and the soil favour the survival in cultivation of a large number of landraces specifically adapted to local conditions. Knowledge on the level and distribution of their genetic variation can help to develop appropriate strategies, in order to suistainably manage in situ these germplasm resources at risk of genetic erosion. C. annuum is an herbaceous diploid species and is considered to be self-pollinating, although different rates of out-crossing have been recorded. We used random amplified polymorphic DNA (RAPD) and amplified fragment length polymorphism (AFLP) markers to assess genetic diversity within and between five populations of a landrace of Capsicum annuum L., grown in a limited area in north-west Italy and locally known as ‘Cuneo’ pepper. Partitioning the genetic variation with Shannon's diversity index revealed that 41.6% occurred between and 58.4% within populations. Analogous results were obtained when the analysis was based only on RAPD or AFLP markers. However, AFLP was more reliable, since a lower range of variation was observed among primer combinations in detecting the two components of genetic variation. Notwithstanding the rather high level of within genetic variation detected, the five populations were clearly differentiated and differed in the frequency of alleles exclusive and/or present at very low frequencies. Our results show the need for accurate estimation of allele frequencies, in order to identify populations to which priority should be given for dynamic conservation of landraces.

Assessment and management of single nucleotide polymorphism genotype errors in genetic association analysis.

Single nucleotide polymorphisms (SNP) may be used in case-control designs to test for association between a marker (the SNP) and a disease. However, such designs usually assume that the genotype data are reported without error. We propose a method, the reduced penetrance model method (RPM) that allows for errors in a case-control design, as compared to the full penetrance model method (FPM), that assumes data are errorless. Pearson's chi 2 applied to a 2 x 2 contingency table is the test statistic considered. Additionally, we provide a likelihood method to estimate error rates using SNP genotype data in CEPH pedigrees. We test our method (RPM) against the standard method (FPM) using simulated data. All SNP loci are assumed to have two alleles, coded 1 and 2. We consider three pairs of error rates, two different sample sizes, and two sets of allele frequencies for the SNP locus. SNP genotype data in two populations are simulated under a null hypothesis (allele frequencies equal in both populations) and under an alternative hypothesis (allele frequencies differ between two populations). The total number of simulations is 24; 12 simulations under the null hypothesis, and 12 simulations under the alternative. The significance level threshold is 5%. For the null case, 9/12 (75%) of the simulations show no increase in type I error under RPM, while 3/12 (25%) show a slight increase (rejecting the null for at most 7% of the replicates). There is no increase in the type I error rate for FPM method, which can also be shown analytically. For the alternative case (power), there is a consistent increase in power for the RPM method as compared to FPM method, and average increase of 0.02 for the simulations considered. When sample sizes are large there is virtually no difference in power between RPM and FPM methods. Also, the RPM method provides consistently more accurate allele frequency estimates for the various populations. Our likelihood method to estimate error rates with CEPH pedigrees provides good estimates on average. The largest difference between a true error rate and our average estimated error rate is 0.006. However, there is a fair amount of variability in the estimates, suggesting the need for multiple experiments or larger numbers of CEPH pedigrees. Researchers may use the methods presented in this paper to (1) estimate error rates for their automated genotyping process, and (2) allow for such errors in association analyses, thereby increasing power to detect differences between allele frequencies in case and control populations when errors are present.