Academic journal article Genetics

A Multivariate Genome-Wide Association Study of Wing Shape in Drosophila Melanogaster

Academic journal article Genetics

A Multivariate Genome-Wide Association Study of Wing Shape in Drosophila Melanogaster

Article excerpt

UNDERSTANDING the inheritance and evolution of complex traits is an important challenge for geneticists and evolutionary biologists alike. A detailed understanding of how genetic variation affects complex traits is important for the treatment of disease, for our attempts to control the evolution of useful or dangerous organisms, and for understanding and predicting evolution over long timescales. Here, we describe the results of a genome-wide association study (GWAS) of the Drosophila melanogaster wing, a model complex trait. We undertook this study as part of our attempt to understand the evolution of the fly wing.

The quantitative genetics of the wing is relatively well studied (Mezey and Houle 2005; Houle and Fierst 2013), yet many aspects of the evolution of wings over short and long timescales are not consistent with the abundant variation we observe (Houle et al. 2003, 2017; Carter and Houle 2011; Pitchers et al. 2013; Bolstad et al. 2015). This suggests that we need a more detailed understanding of the genotypephenotype map (Lewontin 1974), the relationship between genetic variation and the phenotype, to understand the inheritance and evolution of the wing. Fortunately, the fly wing is also a model system for the study of developmental genetics (e.g., de Celis and Diaz-Benjumea 2003; Blair 2007; Wartlicketal. 2011; Matamoro-Vidaletal. 2015), suggesting that the genetic variation influencing the wing can be fitted into a causal framework, directing our attention to the critical aspects of development that enable and shape evolution of the wing. We seek to generate a more precise characterization of the natural genetic variation influencing wing shape than has been possible with previous mapping studies, which utilized far fewer markers than current methods allow (Weber et al. 1999, 2001; Zimmerman et al. 2000; Palsson et al. 2004; Dworkin et al. 2005; Mezey et al. 2005; Dworkin and Gibson 2006).

The evolutionary patterns that we seek to explain concern the relationship of different parts of the wing, rather than the presence or magnitude of single traits. We can measure many aspects of wing shape (Houle et al. 2003), but these are interrelated due to the connections among the cells that make up the wing during development and in the adult. Any developmental change that affects one aspect of the wing, such as the length of a particular vein, must also affect adjacent areas of the wing; any one wing measurement incompletely captures wing variation (Mezey and Houle 2005). This reality of wing morphology is a challenge for association analyses because forward genetic analyses are generally built on analyses of single traits.

Despite growing enthusiasm for a comprehensive phenomic approach (Houle 2010; Houle et al. 2010), the majority of GWASs that include more than one trait use multiple, univariate analyses for each site, rather than a single multivariate analysis (e.g., Teslovich et al. 2010; Battle et al. 2014). Statisticians have long appreciated the value of genuinely multivariate approaches to association studies (Lange et al. 2003; Shriner 2012), leading to a recent proliferation of multivariate methods and software (O'Reilly et al. 2012; Stephens 2013; van der Sluis et al. 2013; Scutari et al. 2014; Zhou and Stephens 2014; Schaid et al. 2016; Porter and O'Reilly 2017). While these methods are diverse, a consistent result is that multivariate analyses increase power to detect associations and the biological usefulness of the results (Porter and O'Reilly 2017). Given these advantages, it is unfortunate that just a few genuinely multivariate empirical association studies have been published (e.g., Anderson et al. 2011; Topp et al. 2013). The majority of published multivariate analyses are examples in the method development papers, instead of fully realized studies.

In this paper, we apply a fully integrated multivariate analysis of the fly wing, drawing on genotypes in the Drosophila Genome Reference Panel (DGRP; Mackay et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.