Pedigree-free animal models: The relatedness matrix reloaded

Article (PDF Available)inProceedings of the Royal Society B: Biological Sciences 275(1635):639-47 · April 2008with41 Reads
DOI: 10.1098/rspb.2007.1032 · Source: PubMed
Animal models typically require a known genetic pedigree to estimate quantitative genetic parameters. Here we test whether animal models can alternatively be based on estimates of relatedness derived entirely from molecular marker data. Our case study is the morphology of a wild bird population, for which we report estimates of the genetic variance-covariance matrices (G) of six morphological traits using three methods: the traditional animal model; a molecular marker-based approach to estimate heritability based on Ritland's pairwise regression method; and a new approach using a molecular genealogy arranged in a relatedness matrix (R) to replace the pedigree in an animal model. Using the traditional animal model, we found significant genetic variance for all six traits and positive genetic covariance among traits. The pairwise regression method did not return reliable estimates of quantitative genetic parameters in this population, with estimates of genetic variance and covariance typically being very small or negative. In contrast, we found mixed evidence for the use of the pedigree-free animal model. Similar to the pairwise regression method, the pedigree-free approach performed poorly when the full-rank R matrix based on the molecular genealogy was employed. However, performance improved substantially when we reduced the dimensionality of the R matrix in order to maximize the signal to noise ratio. Using reduced-rank R matrices generated estimates of genetic variance that were much closer to those from the traditional model. Nevertheless, this method was less reliable at estimating covariances, which were often estimated to be negative. Taken together, these results suggest that pedigree-free animal models can recover quantitative genetic information, although the signal remains relatively weak. It remains to be determined whether this problem can be overcome by the use of a more powerful battery of molecular markers and improved methods for reconstructing genealogies.
    • "However, to date the pedigree approach has been preferred in wild populations because estimating genome-wide relatedness accurately requires a very large number of markers. In particular, methods developed to take advantage of the small microsatellite datasets typical of molecular ecology studies over the last two decades tend to be statistically very 'noisy' (Ritland, 1996; Wilson et al., 2003; Frentiu et al., 2008). Note however that this issue can now be overcome by using very high densities of SNP markers (Bérénos et al., 2014) and we anticipate a large growth in such 'pedigree-free' quantitative genetic studies in the very near future as a consequence of the adoption of next generation sequencing technologies. "
    [Show abstract] [Hide abstract] ABSTRACT: Phenotypes evolve under natural selection if, and only if, they are genetically variable. While evolutionary ecologists have long studied natural selection, it is only comparatively recently that quantitative genetic methods have been applied to wild populations. This rapidly growing area of research is allowing us to scrutinize the genetic basis of (co)variation within- and among-traits, increasing our understanding of adaptive evolution in nature. Here we review some of the key principles, developments, challenges, and emerging directions in this rapidly growing field of research.
    Full-text · Article · Dec 2016 · Ecology and Evolution
    • "We proposed a modification of the matrix that solved the problem adequately (without this approximation the biases ranged from À10 to À55% in scenario i). Other methods have been suggested to deal with noninversible matrices, which is a classical issue for pedigreefree methods (Frentiu et al. 2008; Ahlinder & Sillanpaa 2013 ). Because of its robustness to various ecological scenarios , we suggest to use the 'Hybrid' approach in sibling quantitative genetic designs for populations where no a priori knowledge exists about the mating system or the determinism of the traits. "
    [Show abstract] [Hide abstract] ABSTRACT: Accurate estimates of heritability (h(2) ) are necessary to assess adaptive responses of populations and evolution of fitness-related traits in changing environments. For plants, h(2) estimates generally rely on maternal progeny designs, assuming that offspring are either half-sibs or unrelated. However, plant mating systems often depart from half-sib assumptions, this can bias h(2) estimates. Here, we investigate how to accurately estimate h(2) in non-model species through the analysis of sibling designs with a moderate genotyping effort. We performed simulations to investigate how microsatellite marker information available for only a subset of offspring can improve h(2) estimates based on maternal progeny designs in presence of non-random mating, inbreeding in the parental population or maternal effects. We compared the basic family method, considering or not adjustments based on average relatedness coefficients, and methods based on the animal model. The animal model was used with average relatedness information, or with hybrid relatedness information: associating one-generation pedigree and family assumptions, or associating one-generation pedigree and average relatedness coefficients. Our results highlighted that methods using marker-based relatedness coefficients performed as well as pedigree-based methods in presence of non-random mating (i.e. unequal male reproductive contributions, selfing), offering promising prospects to investigate in situ heritabilities in natural populations. In presence of maternal effects, only the use of pairwise relatednesses through pedigree information improved the accuracy of h(2) estimates. In that case the amount of father-related offspring in the sibling design is the most critical. Overall, we showed that the method using both one-generation pedigree and average relatedness coefficients was the most robust to various ecological scenarios. This article is protected by copyright. All rights reserved.
    Full-text · Article · Jun 2016
    • "Estimating heritability requires a known pedigree or marker-inferred pairwise relatedness between individuals, and quantitative trait values of these individuals. One powerful approach for estimating heritabilities in the wild is to use these pairwise relatedness estimates in so-called animal-model analyses (Milner et al. 2000; Frentiu et al. 2008; B er enos et al. 2014; Kl ap st e et al. 2014 ). Animalmodel analyses are linear mixed models that can be used to estimate the additive genetic contribution to phenotypic trait variation and thus trait heritability (Wilson et al. 2010). "
    [Show abstract] [Hide abstract] ABSTRACT: Alpine ecosystems are seriously threatened by climate change. One of the key mechanisms by which plants can adapt to changing environmental conditions is through evolutionary change. However, we still know little about the evolutionary potential in wild populations of long-lived alpine plants. Here, we investigated heritabilities of phenological traits, leaf size, and performance traits in natural populations of the long-lived alpine dwarf shrub Salix herbacea using relatedness estimates inferred from SSR (Simple Sequence Repeat) markers. Salix herbacea occurs in early- and late-snowmelt microhabitats (ridges and snowbeds), and we assessed how performance consequences of phenological traits and leaf size differ between these microhabitats in order to infer potential for evolutionary responses. Salix herbacea showed low, but significant, heritabilities of leaf size, clonal and sexual reproduction, and moderate heritabilities of phenological traits. In both microhabitats, we found that larger leaves, longer intervals between snowmelt and leaf expansion, and longer GDD (growing-degree days) until leaf expansion resulted in a stronger increase in the number of stems (clonal reproduction). In snowbeds, clonal reproduction increased with a shorter GDD until flowering, while the opposite was found on ridges. Furthermore, the proportion of flowering stems increased with GDD until flowering in both microhabitats. Our results suggest that the presence of significant heritable variation in morphology and phenology might help S. herbacea to adapt to changing environmental conditions. However, it remains to be seen if the rate of such an evolutionary response can keep pace with the rapid rate of climate change.
    Full-text · Article · May 2016
Show more