Mapping Of Complex Traits Research Articles

Widespread Allelic Heterogeneity in Complex Traits

Recent successes in genome-wide association studies (GWASs) make it possible to address important questions about the genetic architecture of complex traits, such as allele frequency and effect size. One lesser-known aspect of complex traits is the extent of allelic heterogeneity (AH) arising from multiple causal variants at a locus. We developed a computational method to infer the probability of AHand applied it to three GWASs and four expression quantitative trait loci (eQTL) datasets. We identified a total of 4,152 loci with strong evidence of AH. The proportion of all loci with identified AH is 4%-23% in eQTLs, 35% in GWASs of high-density lipoprotein (HDL), and 23% in GWASs of schizophrenia. For eQTLs, we observed a strong correlation between sample size and the proportion of loci with AH (R2 = 0.85, p = 2.2× 10-16), indicating that statistical power prevents identification of AH in other loci. Understanding the extent of AH may guide the development of new methods for fine mapping and association mapping of complex traits.

A nonparametric approach for functional mapping of complex traits

A nonparametric approach for functional mapping of complex traits

Mapping Complex Traits in a Diversity Outbred F1 Mouse Population Identifies Germline Modifiers of Metastasis in Human Prostate Cancer

It is unclear how standing genetic variation affects the prognosis of prostate cancer patients. To provide one controlled answer to this problem, we crossed a dominant, penetrant mouse model of prostate cancer to Diversity Outbred mice, a collection of animals that carries over 40 million SNPs. Integration of disease phenotype and SNP variation data in 493 F1 males identified a metastasis modifier locus on Chromosome 8 (LOD= 8.42); further analysis identified thegenes Rwdd4, Cenpu, and Casp3 as functional effectors of this locus. Accordingly, analysis of over 5,300 prostate cancer patient samples revealed correlations between the presence of genetic variants atthese loci, their expression levels, cancer aggressiveness, and patient survival. We also observed that ectopic overexpression of RWDD4 and CENPU increased the aggressiveness of two human prostate cancer cell lines. In aggregate, our approach demonstrates how well-characterized genetic variation in mice can be harnessed in conjunction with systems genetics approaches to identify and characterize germline modifiers of human disease processes.

Bioinformatic analysis of genotype by sequencing (GBS) data with NGSEP

BackgroundTherecent development and availability of different genotype by sequencing (GBS) protocols provided a cost-effective approach to perform high-resolution genomic analysis of entire populations in different species. The central component of all these protocols is the digestion of the initial DNA with known restriction enzymes, to generate sequencing fragments at predictable and reproducible sites. This allows to genotype thousands of genetic markers on populations with hundreds of individuals. Because GBS protocols achieve parallel genotyping through high throughput sequencing (HTS), every GBS protocol must include a bioinformatics pipeline for analysis of HTS data. Our bioinformatics group recently developed the Next Generation Sequencing Eclipse Plugin (NGSEP) for accurate, efficient, and user-friendly analysis of HTS data.ResultsHere we present the latest functionalities implemented in NGSEP in the context of the analysis of GBS data. We implemented a one step wizard to perform parallel read alignment, variants identification and genotyping from HTS reads sequenced from entire populations. We added different filters for variants, samples and genotype calls as well as calculation of summary statistics overall and per sample, and diversity statistics per site. NGSEP includes a module to translate genotype calls to some of the most widely used input formats for integration with several tools to perform downstream analyses such as population structure analysis, construction of genetic maps, genetic mapping of complex traits and phenotype prediction for genomic selection. We assessed the accuracy of NGSEP on two highly heterozygous F1 cassava populations and on an inbred common bean population, and we showed that NGSEP provides similar or better accuracy compared to other widely used software packages for variants detection such as GATK, Samtools and Tassel.ConclusionsNGSEP is a powerful, accurate and efficient bioinformatics software tool for analysis of HTS data, and also one of the best bioinformatic packages to facilitate the analysis and to maximize the genomic variability information that can be obtained from GBS experiments for population genomics.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-016-2827-7) contains supplementary material, which is available to authorized users.

Solanum pennellii backcross inbred lines (BILs) link small genomic bins with tomato traits.

We present a resource for fine mapping of traits derived from the wild tomato species Solanum pennellii (LA0716). The population of backcross inbred lines (BILs) is composed of 446 lines derived after a few generations of backcrosses of the wild species with cultivated tomato (cultivar M82; LA3475), followed by more than seven generations of self-pollination. The BILs were genotyped using the 10K SOL-CAP single nucleotide polymorphism (SNP) -Chip, and 3700 polymorphic markers were used to map recombination break points relative to the physical map of Solanum lycopersicum. The BILs carry, on average, 2.7 introgressions per line, with a mean introgression length of 11.7Mbp. Whereas the classic 76 introgression lines (ILs) partitioned the genome into 106 mapping bins, the BILs generated 633 bins, thereby enhancing the mapping resolution of traits derived from the wild species. We demonstrate the power of the BILs for rapid fine mapping of simple and complex traits derived from the wild tomato species.

Assessing the value of phenotypic information from non-genotyped animals for QTL mapping of complex traits in real and simulated populations.

BackgroundQTL mapping through genome-wide association studies (GWAS) is challenging, especially in the case of low heritability complex traits and when few animals possess genotypic and phenotypic information. When most of the phenotypic information is from non-genotyped animals, GWAS can be performed using the weighted single-step GBLUP (WssGBLUP) method, which permits to combine all available information, even that of non-genotyped animals. However, it is not clear to what extent phenotypic information from non-genotyped animals increases the power of QTL detection, and whether factors such as the extent of linkage disequilibrium (LD) in the population and weighting SNPs in WssGBLUP affect the importance of using information from non-genotyped animals in GWAS. These questions were investigated in this study using real and simulated data.ResultsAnalysis of real data showed that the use of phenotypes of non-genotyped animals affected SNP effect estimates and, consequently, QTL mapping. Despite some coincidence, the most important genomic regions identified by the analyses, either using or ignoring phenotypes of non-genotyped animals, were not the same. The simulation results indicated that the inclusion of all available phenotypic information, even that of non-genotyped animals, tends to improve QTL detection for low heritability complex traits. For populations with low levels of LD, this trend of improvement was less pronounced. Stronger shrinkage on SNPs explaining lower variance was not necessarily associated with better QTL mapping.ConclusionsThe use of phenotypic information from non-genotyped animals in GWAS may improve the ability to detect QTL for low heritability complex traits, especially in populations in which the level of LD is high.

A time-varying group sparse additive model for genome-wide association studies of dynamic complex traits.

Despite the widespread popularity of genome-wide association studies (GWAS) for genetic mapping of complex traits, most existing GWAS methodologies are still limited to the use of static phenotypes measured at a single time point. In this work, we propose a new method for association mapping that considers dynamic phenotypes measured at a sequence of time points. Our approach relies on the use of Time-Varying Group Sparse Additive Models (TV-GroupSpAM) for high-dimensional, functional regression. This new model detects a sparse set of genomic loci that are associated with trait dynamics, and demonstrates increased statistical power over existing methods. We evaluate our method via experiments on synthetic data and perform a proof-of-concept analysis for detecting single nucleotide polymorphisms associated with two phenotypes used to assess asthma severity: forced vital capacity, a sensitive measure of airway obstruction and bronchodilator response, which measures lung response to bronchodilator drugs. Source code for TV-GroupSpAM freely available for download at http://www.cs.cmu.edu/~mmarchet/projects/tv_group_spam, implemented in MATLAB. epxing@cs.cmu.edu Supplementary data are available at Bioinformatics online.

TOPOLOGICAL METHODS FOR THE QUANTIFICATION AND ANALYSIS OF COMPLEX PHENOTYPES

Quantitative Trait Locus (QTL) mapping of complex traits, such as leaf venation or root structures, require the phenotyping and genotyping of large populations. Sufficient genotyping is accomplished with cost effective high-throughput assays, however labor costs often makes sufficient phenotyping prohibitively limited. In order to develop efficient high-throughput phenotyping platforms for complex traits algorithms and methods for quantifying these traits are needed. It is often desirable to study the spatial organization of these phenotypes from the images generated by high-throughput platforms. With the goal of quantifying the traits, many approaches try to identify several core traits useful in describing the phenotypic morphology. This simplification may lose important information about the phenotype. Rather than reducing the structural information, we introduce a novel method, the Persistence Intensity Array, for studying complex traits using tools from the emergent field of Topological Data Analysis. This approach uses the complete geometry of the phenotype and represents it as a simpler summary of the key topological shape features contained in the data. We demonstrate this method's efficacy by through a simulated QTL analysis.

Target Enrichment Improves Mapping of Complex Traits by Deep Sequencing.

Complex traits such as crop performance and human diseases are controlled by multiple genetic loci, many of which have small effects and often go undetected by traditional quantitative trait locus (QTL) mapping. Recently, bulked segregant analysis with large F2 pools and genome-level markers (named extreme-QTL or X-QTL mapping) has been used to identify many QTL. To estimate parameters impacting QTL detection for X-QTL mapping, we simulated the effects of population size, marker density, and sequencing depth of markers on QTL detectability for traits with differing heritabilities. These simulations indicate that a high (>90%) chance of detecting QTL with at least 5% effect requires 5000× sequencing depth for a trait with heritability of 0.4−0.7. For most eukaryotic organisms, whole-genome sequencing at this depth is not economically feasible. Therefore, we tested and confirmed the feasibility of applying deep sequencing of target-enriched markers for X-QTL mapping. We used two traits in Arabidopsis thaliana with different heritabilities: seed size (H2 = 0.61) and seedling greening in response to salt (H2 = 0.94). We used a modified G test to identify QTL regions and developed a model-based statistical framework to resolve individual peaks by incorporating recombination rates. Multiple QTL were identified for both traits, including previously undiscovered QTL. We call our method target-enriched X-QTL (TEX-QTL) mapping; this mapping approach is not limited by the genome size or the availability of recombinant inbred populations and should be applicable to many organisms and traits.

Construction of the High-Density Genetic Linkage Map and Chromosome Map of Large Yellow Croaker (Larimichthys crocea).

High-density genetic maps are essential for genome assembly, comparative genomic analysis and fine mapping of complex traits. In this study, 31,191 single nucleotide polymorphisms (SNPs) evenly distributed across the large yellow croaker (Larimichthys crocea) genome were identified using restriction-site associated DNA sequencing (RAD-seq). Among them, 10,150 high-confidence SNPs were assigned to 24 consensus linkage groups (LGs). The total length of the genetic linkage map was 5451.3 cM with an average distance of 0.54 cM between loci. This represents the densest genetic map currently reported for large yellow croaker. Using 2889 SNPs to target specific scaffolds, we assigned 533 scaffolds, comprising 421.44 Mb (62.04%) of the large yellow croaker assembled sequence, to the 24 linkage groups. The mapped assembly scaffolds in large yellow croaker were used for genome synteny analyses against the stickleback (Gasterosteus aculeatus) and medaka (Oryzias latipes). Greater synteny was observed between large yellow croaker and stickleback. This supports the hypothesis that large yellow croaker is more closely related to stickleback than to medaka. Moreover, 1274 immunity-related genes and 195 hypoxia-related genes were mapped to the 24 chromosomes of large yellow croaker. The integration of the high-resolution genetic map and the assembled sequence provides a valuable resource for fine mapping and positional cloning of quantitative trait loci associated with economically important traits in large yellow croaker.

Multiple SNP Set Analysis for Genome-Wide Association Studies Through Bayesian Latent Variable Selection.

The power of genome-wide association studies (GWAS) for mapping complex traits with single-SNP analysis (where SNP is single-nucleotide polymorphism) may be undermined by modest SNP effect sizes, unobserved causal SNPs, correlation among adjacent SNPs, and SNP-SNP interactions. Alternative approaches for testing the association between a single SNP set and individual phenotypes have been shown to be promising for improving the power of GWAS. We propose a Bayesian latent variable selection (BLVS) method to simultaneously model the joint association mapping between a large number of SNP sets and complex traits. Compared with single SNP set analysis, such joint association mapping not only accounts for the correlation among SNP sets but also is capable of detecting causal SNP sets that are marginally uncorrelated with traits. The spike-and-slab prior assigned to the effects of SNP sets can greatly reduce the dimension of effective SNP sets, while speeding up computation. An efficient Markov chain Monte Carlo algorithm is developed. Simulations demonstrate that BLVS outperforms several competing variable selection methods in some important scenarios.

Abstract 4138: Integrating SNPs, epigenetics and transcriptomics to better understand the inherited predisposition to breast cancer metastasis

Abstract Breast cancer is the most frequent cancer and the second leading cause of cancer mortality in women. The vast majority of breast cancer mortality is due to metastatic disease since the primary tumor can be relatively easily surgically resected. More comprehensive understanding of biology of metastasis is therefore clearly warranted to unveil novel metastasis-associated molecules and cellular processes that might be targeted for clinical intervention. Previously we demonstrated that like cancer incidence, metastatic progression has a significant inherited component. Using mouse complex trait mapping strategies we have been isolating metastasis susceptibility genes from backcross analysis. However, studies suggest that many of these QTL peaks are the result of the contribution of multiple genes, suggesting that causal genes are not being identified. To increase our ability to detect causal genes we have implemented a new integrated strategy. Whole genome sequencing of appropriate high or low metastatic mouse strains is performed to identify variants, which are then filtered for those within DNAse hypersensitive sites (DHS), based on the hypothesis that most inherited phenotypes are due to expression differences rather than missense variants. The genes associated with polymorphic DHS are then screened through mouse and human tumor expression databases to identify those genes associated with development of metastatic disease. This strategy was able to identify genes previously associated with metastatic breast cancer in both mouse and humans and new genes have also been validated. The results further suggest that inherited variation in cellular mediated immune response may be an important contributor to metastatic disease. Citation Format: Anjali Shukla, Ling Bai, Howard Yang, Anthony Doran, Ying Hu, Thomas Geiger, Maxwell Lee, Thomas Keane, Kent W. Hunter. Integrating SNPs, epigenetics and transcriptomics to better understand the inherited predisposition to breast cancer metastasis. [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 4138. doi:10.1158/1538-7445.AM2015-4138

Informatics resources for the Collaborative Cross and related mouse populations.

Informatics resources for the Collaborative Cross and related mouse populations.

Admixture mapping of tuberculosis and pigmentation-related traits in an African-European hybrid cattle population.

Admixture mapping affords a powerful approach to genetic mapping of complex traits and may be particularly suited to investigation in cattle where many breeds and populations are hybrids of the two divergent ancestral genomes, derived from Bos taurus and Bos indicus. Here we design a minimal genome wide SNP panel for tracking ancestry in recent hybrids of Holstein–Friesian and local Arsi zebu in a field sample from a region of high bovine tuberculosis (BTB) endemicity in the central Ethiopian highlands. We first demonstrate the utility of this approach by mapping the red coat color phenotype, uncovering a highly significant peak over the MC1R gene and a second peak with no previously known candidate gene. Secondly, we exploit the described differential susceptibility to BTB between the ancestral strains to identify a region in which Bos taurus ancestry associates, at suggestive significance, with skin test positivity. Interestingly, this association peak contains the toll-like receptor gene cluster on chromosome 6. With this work we have shown the potential of admixture mapping in hybrid domestic animals with divergent ancestral genomes, a recurring condition in domesticated species.

Toward the practical utility of systems mapping: Reply to comments on “Mapping complex traits as a dynamic system”

Toward the practical utility of systems mapping: Reply to comments on “Mapping complex traits as a dynamic system”

Systems mapping has potential to overcome inherent problems of genetic mapping: Comment on “Mapping complex traits as a dynamic system” by L. Sun and R. Wu

The development of a zygote into an adult organism has puzzled scientists for a long time. The degree to which developmental processes are subject to genetic control – i.e., the degree to which adult traits are heritable – has been estimated by quantitative geneticists since the early 20th century [9]. Only in the second half of the 20th century, after the discovery of the structure of the DNA, the molecular mechanisms of inheritance have begun to be understood. The development of effective techniques for screening DNA polymorphisms allowed for statistical studies of the association of quantitative traits with genetic markers [8]. Since then, a suite of quantitative methods has been established for investigating statistical genotype–phenotype associations, which have been applied extensively in medical contexts as well [2,3,10]. Simultaneously, and largely independently from these developments, mathematical approaches have been devised to model the interaction of molecular components, such as enzyme kinetics, within and across cells. The field of systems biology attempts to model the mechanistic relationships between a multitude of molecular components in the developing and adult organism [7,14]. The relatively new concept of systems mapping [4,5,15], excellently reviewed by Sun and Wu [13], builds a bride between these two – hitherto largely unconnected – traditions in the biosciences and offers potential for the merging of different lines of empirical and theoretical evidence. At the core of systems mapping is a set of ordinary differential equations (ODEs) that describe the dynamics of a biological system. Sun and Wu present multiple examples of such systems, including growth rates in plants, pharmacodynamic reactions, and phenotype landscapes. The individual parameters of the ODEs are then used as variables for the genetic mapping. One inherent and often neglected ambiguity of classic genetic mappings is their dependence on the scale of the variables. For instance, the effect of an allele on a given quantitative trait might comprise both additive and non-additive (dominance) components. But under certain non-linear transformations of the phenotypic variable (such as raising it to some power or applying a log transform), the effect might change to a purely additive one, without a dominance

How to dissect complex traits and how to choose suitable mapping resources for system genetics?: Comment on “Mapping complex traits as a dynamic system” by L. Sun and R. Wu

How to dissect complex traits and how to choose suitable mapping resources for system genetics?: Comment on “Mapping complex traits as a dynamic system” by L. Sun and R. Wu

Leveraging Multi-Ethnic Evidence for Mapping Complex Traits in Minority Populations: An Empirical Bayes Approach

Elucidating the genetic basis of complex traits and diseases in non-European populations is particularly challenging because U.S. minority populations have been under-represented in genetic association studies. We developed an empirical Bayes approach named XPEB (Cross-Population Empirical Bayes), designed to improve the power for mapping complex trait loci in a minority population by exploiting information from genome-wide association studies (GWAS) from another ethnic population. Taking as input summary statistics from two GWAS, a target-GWAS from an ethnic minority population of primary interest and an auxiliary base-GWAS (such as a larger GWAS in Europeans), our XPEB approach reprioritizes SNPs in the target population to compute local false discovery rates. We demonstrated, through simulations, that whenever the base-GWAS harbors relevant information, XPEB gains efficiency. Moreover, XPEB has the ability to discard irrelevant auxiliary information, providing a safeguard against inflated false discovery rates due to genetic heterogeneity between populations. Applied to a blood lipids study in African Americans, XPEB more than quadrupled the discoveries compared to the conventional approach that uses target-GWAS alone, bringing the number of significant loci from 14 to 65. Thus, XPEB offers a flexible framework for mapping complex traits in minority populations.

Ultrahigh-density linkage map for cultivated cucumber (Cucumis sativus L.) using a single-nucleotide polymorphism genotyping array.

Genotyping arrays are tools for high-throughput genotyping, which is beneficial in constructing saturated genetic maps and therefore high-resolution mapping of complex traits. Since the report of the first cucumber genome draft, genetic maps have been constructed mainly based on simple-sequence repeats (SSRs) or on combinations of SSRs and sequence-related amplified polymorphism (SRAP). In this study, we developed the first cucumber genotyping array consisting of 32,864 single-nucleotide polymorphisms (SNPs). These markers cover the cucumber genome with a median interval of ~2 Kb and have expected genotype calls in parents/F1 hybridizations as a training set. The training set was validated with Fluidigm technology and showed 96% concordance with the genotype calls in the parents/F1 hybridizations. Application of the genotyping array was illustrated by constructing a 598.7 cM genetic map based on a ‘9930’ × ‘Gy14’ recombinant inbred line (RIL) population comprised of 11,156 SNPs. Marker collinearity between the genetic map and reference genomes of the two parents was estimated at R2 = 0.97. We also used the array-derived genetic map to investigate chromosomal rearrangements, regional recombination rate, and specific regions with segregation distortions. Finally, 82% of the linkage-map bins were polymorphic in other cucumber variants, suggesting that the array can be applied for genotyping in other lines. The genotyping array presented here, together with the genotype calls of the parents/F1 hybridizations as a training set, should be a powerful tool in future studies with high-throughput cucumber genotyping. An ultrahigh-density linkage map constructed by this genotyping array on RIL population may be invaluable for assembly improvement, and for mapping important cucumber QTLs.

QTL mapping – Current status and challenges: Comment on “Mapping complex traits as a dynamic system” by L. Sun and R. Wu

QTL mapping – Current status and challenges: Comment on “Mapping complex traits as a dynamic system” by L. Sun and R. Wu