This project’s overall goal is to help close the gap between the increasingly rapid ability to sequence the DNA (genotype) of an organism and the relative lack of techniques for measuring the even more complex physical form (phenotype) of the organism. The measurement of complex, high-dimensional phenotypes is an essential complement to the genomic level data. We call the integration of such phenotypic data with genetics “phenomics”. This project exploits an existing system for automated wing measurement, and will develop techniques for the rapid measurement of other phenotypes, including whole flies. From a biological point of view, the overall aim of this project is to understand the genetic causes of phenotypic variation in the wings and other aspects of external morphology of Drosophila melanogaster. The proposed work will test the usefulness of a phenomic approach for the Drosophila wing.
The overall aim of this project is to understand the genetic causes of phenotypic variation in the wings and other aspects of external morphology of Drosophila melanogaster. More broadly, we wish to develop and validate an approach to the study of the relationship between genotype and phenotype that we call phenomic. The key idea is that the measurement of complex, high-dimensional phenotypes is an essential complement to the genomic level data that is increasingly easy to obtain. Many of the important problems in biology require us to determine the relationship between the genotype and the phenotype; we will not make progress on this until we treat the measurement of the phenotype as an important task. In the proposed work, we wish to test the usefulness of this approach for the Drosophila wing, and to develop capabilities to rapidly measure other aspects of external morphology. Our specific aims are summed up in three questions:
- Aim 1: What genetic changes can cause variation in wing shape? We will develop a “dictionary” of genetic effects on wing form by systematically manipulating gene expression at genetic loci hypothesized to be involved in wing development. The PI’s semi-automated system for measuring wing phenotypes will be used in conjunction with the genetic resources of D. melanogaster to rapidly characterize the phenotypic effects at a wide range of genes. We will engineer novel variants in a few key genes to characterize the sensitivity of the phenotype to perturbations in gene expression with known timing and magnitude. Once these data are in hand, we will then summarize and model these data to answer the following questions: Are all of the genes in a particular developmental pathway producing a similar set of transformations of form when mutated? Are some kinds of form transformations easier to obtain by perturbation than others? Is there a limited set of possible form transformations? These effects will then be used to build predictive models of the connections between developmental pathways and phenotypic variation.
- Aim 2:: What are the genetic causes of natural variation in wing shape? We will apply the phenomic approach to the genetic basis of natural variation in wing shape in D. melanogaster using the dictionary of genetic effects and the resulting models generated in Specific Aim 1. We will characterize the quantitative genetic variation in wing shape in a natural population of D. melanogaster, then attempt to infer its possible causes based on the correspondence of the observed variation with the dictionary. These inferences will be tested using QTL mapping and quantitative complementation. Both exploratory and confirmatory (structural equation) models will be fit to the data. Finally, we will apply this approach to the PI’s existing data sets on the quantitative genetic variation in wing form in populations of three Drosophila species.
- Aim 3: How do genetic changes that cause variation in wing shape affect other body parts? Pleiotropy, or the effects of a single genetic variant on multiple traits, is an important feature of genetic systems, with profound implications for evolution. It is difficult to study because of the sheer number of possible phenotypic effects that can be involved. We will develop and generalize methods to measure variation in Drosophila eggs and whole flies. A detailed quantitative atlas of Drosophila morphology will first be generated, then used to map 2-D images onto the atlas, allowing quantitative information on 3-D form to be extracted while minimizing human involvement. The pleiotropic effects of the variants characterized in Aim 1 will then be characterized for their pleiotropic effects.