Pleiotropy between Drosophila melanogaster phenotypic modules. Image is modified, with permission, from Ayroles A. J. et al. Nature Genet. (doi: 10.1038/ng.332) © (2009) Macmillan Publishers Ltd. All rights reserved.

Determining the genetic architecture of complex traits poses a challenge because most phenotypic variation is caused by numerous interactions between a multitude of environmentally sensitive genes. A recent study that uses systems genetics to analyse complex traits in Drosophila melanogaster is an important step towards understanding the complex genetic networks that underpin ecologically relevant phenotypes in this organism.

Ayroles et al. used microarrays to quantify genome-wide transcript abundance in a resource of 40 inbred D. melanogaster lines that were derived from the wild. They found a much greater degree of genetic variation than had been uncovered by previous studies: almost 80% of the D. melanogaster genome (∼10,000 transcripts) was expressed in adult flies, and two-thirds of the expressed transcripts were genetically variable.

There was a high degree of correlation between the variably expressed transcripts. The authors were able to group these transcripts into 241 modules, with each module consisting of a separate cluster of highly interconnected genes. The transcripts in a given module were often biologically related; for example, modules could be enriched for transcripts that function in a common pathway, or for transcripts that show a similar tissue-specific pattern of expression.

The authors found high levels of genetic variation between the fly lines for a number of ecologically relevant traits that included resistance to starvation, lifespan and mating speed. They discovered several hundred transcripts and single feature polymorphisms (SFPs) that associated with the phenotypic variation in each quantitative trait, allowing them to identify candidate genes for each trait. These candidate genes were then validated by analysing the phenotypes of P element insertion mutations in or near the gene of interest. The authors identified distinct transcriptional modules that contained transcripts associated with each complex trait. For example, transcripts associated with variation for fitness were enriched for genes that mediate functions such as the immune response, chemosensation and visual perception.

The systems genetics approach used in this study is a powerful method for identifying novel candidate genes for complex traits and for determining the relationships between these genes. Notably, most of the candidate transcripts identified were unexpected based on previous mutational analyses of the traits. Moreover, the inbred fly lines and transcriptional data generated from the study are a valuable resource for the Drosophila community as they can be used to analyse other complex phenotypes.