Mackay-Anholt lab

The Genetic Basis of Quantitative Variation

In contrast to Mendelian traits, which are driven by expression of a single gene, variation in most traits among individuals in a population arises from the coordinated expression of many genes, environmental effects, and the interplay between the environment and the genome. Such traits are known as “quantitative traits” and they encompass morphological traits (e.g. height), physiological traits (e.g. blood pressure), behavioral traits (e.g. propensity for addiction), life history traits (e.g. life span) and risk for common diseases (e.g. cardiovascular disease, diabetes and cancer).  Because of the complex underpinnings of quantitative traits and the interplay between “nature” and “nurture” there is no linear relationship between genetic variation and variation in the organismal phenotype.

Our goal for understanding the genetic architecture of quantitative traits is to elucidate the rules for translating genetic variation among individuals to phenotypic variation for the trait, at the level of primary variation in DNA sequence and intermediate phenotypes of transcript, protein and metabolite abundance, and in a range of relevant environments.

Drosophila melanogaster as a model for studies on complex traits  

The fruitfly, Drosophila melanogaster, provides a powerful model system for uncovering fundamental principles of the genotype-phenotype relationship of quantitative traits which are relevant to advancing our knowledge of the contributions of gene-gene interactions and environmental conditions to human traits for health and disease. Flies have a short generation time and can be grown in large numbers, economically and without regulatory restrictions under controlled environmental conditions and with defined genetic backgrounds. They are readily amenable to genetic manipulations and display a wide range of complex behavioral morphological and physiological traits. We can use mutational analysis, RNA interference, CRISPR-mediated gene deletion and gene editing, analyses of chromatin structure and epigenetic modifications, as well as classical gene mapping approaches to identify genes and genetic networks associated with variation in a wide range of complex traits.

Genome-wide association analyses have become available with the generation of the Drosophila melanogaster Genetic Reference Panel (DGRP), which was constructed in our laboratories as a community resource. The DGRP consists of 205 wild-derived inbred lines with fully sequenced, well annotated genomes. Whereas genetic variation among the lines reflects the variation in the natural population from which the lines were derived, individuals within each line retain minimal genetic variation; Thus, the same genotypes can be measured repeatedly for the same trait or different traits and in different environments.   Thus, the same strains can be evaluated for multiple complex traits, including ‘intermediate’ phenotypes such as whole genome transcript abundance and quantitative variation in the proteome and metabolome. This facilitates a systems genetics approach for understanding the genetic architecture of complex traits in an economical genetic model organism.

A sample of 205 strains has the power to detect intermediate frequency variants with moderately small to large effects on complex traits. A user-friendly pipeline to map variants associated with measured quantitative traits using a sophisticated mixed linear model to account for effects of Wolbachia infection status, karyotype of common polymorphic inversions, and cryptic relatedness is available at dgrp2.gnets.ncsu.edu.

The limited power to detect alleles with smaller effects in the original DGRP lines can be overcome by constructing advanced intercross populations from subsets of DGRP lines, either chosen at random or from lines that are divergent for the trait of interest. The populations can be subjected to extreme QTL mapping designs (bulk segregant analyses) to rapidly and precisely map variants affecting complex traits; confirm additive effects of variants identified by association mapping in the DGRP; identify variants that have epistatic gene action in the DGRP; and used for laboratory evolution studies, including artificial selection.

Drosophila Models of Human Disease

Approximately 67% of Drosophila genes have a human ortholog, and 75% of human disease genes have a Drosophila ortholog. Common and rare genetic diseases and disorders affect a large fraction of the world’s population. Common diseases and risk factors such as hypertension, high cholesterol, arthritis, heart disease and diabetes collectively affect ~80% of the population, with many individuals afflicted by more than one disease. These traits have a complex genetic basis due to segregating alleles with small effects at many loci and are also affected by environmental exposures: they are quantitative traits. In addition to common diseases ~7,000 rare diseases have been described. Although each rare disease affects less than 1 in 200,000 individuals; collectively, rare diseases affect up to 10% of live births, with a large pediatric patient population. Most rare diseases are caused by a genetic disruption with a highly penetrant effect that segregates in pedigrees as a Mendelian polymorphism. Mapping the causal genetic basis of common and rare diseases and understanding the mechanism(s) whereby causal variants affect disease are critical for diagnosis, risk prediction, management of patient symptoms and the development of therapies.

We use Drosophila melanogaster to study the genetic basis of a wide variety of traits, including lifespan, cocaine and methamphetamine consumption and preference, food consumption, sensitivity and resistance to the inebriating effects of ethanol, starvation stress resistance, aggressive behavior and locomotor behavior in Drosophila. We use association mapping in the DGRP and Advanced Intercross Populations derived from subsets of DGRP lines to identify candidate genes and naturally occurring variants affecting these traits. We then test whether the candidate genes affect the trait using RNAi. Finally, we are using gene editing to functionally assess the variant effects in multiple DGRP lines. We also assess variation in gene expression in the DGRP lines to derive causal transcriptional networks associated with these quantitative traits. Based on the principle of evolutionary conservation of fundamental biological processes we can extrapolate and apply findings from our work on Drosophila to analogous human disorders (e.g. propensity for substance abuse).

In addition to common disorder we are also using Drosophila to model rare human diseases in two ways. First, we are performing whole genome gene expression analysis of multiple genes known to affect single diseases or syndromes to identify common co-regulated genes; these genes are then novel candidate genes affecting the disease for which clinical tests can be developed for patients who remain undiagnosed. Second, we are using gene editing to replace the Drosophila ortholog of the disease gene with the wild type human gene as well as the known causal mutations in the human gene and performing phenotypic screens (including effects on genome wide gene expression) to assess the extent to which disease symptoms can be modeled in Drosophila. We then proceed to testing the effects of variants of unknown significance that have been identified in patients with the diseases, as well as screening for genetic modifiers using the DGRP. These projects are in collaboration with Dr. Richard Steet and Dr. Heather Flanagan-Steet and clinicians and diagnosticians at the Greenwood Genetic Center.