BACKGROUND:Large scale screening for synthetic lethality serves as a common tool in yeast genetics to systematically search for genes that play a role in specific biological processes. Often the amounts of data resulting from a single large scale screen far exceed the capacities of experimental characterization of every identified target. Thus, there is need for computational tools that select promising candidate genes in order to reduce the number of follow-up experiments to a manageable size.RESULTS:We analyze synthetic lethality data for arp1 and jnm1, two spindle migration genes, in order to identify novel members in this process. To this end, we use an unsupervised statistical method that integrates additional information from biological data sources, such as gene expression, phenotypic profiling, RNA degradation and sequence similarity. Different from existing methods that require large amounts of synthetic lethal data, our method merely relies on synthetic lethality information from two single screens. Using a Multivariate Gaussian Mixture Model, we determine the best subset of features that assign the target genes to two groups. The approach identifies a small group of genes as candidates involved in spindle migration. Experimental testing confirms the majority of our candidates and we present she1 (YBL031W) as a novel gene involved in spindle migration. We applied the statistical methodology also to TOR2 signaling as another example.CONCLUSION:We demonstrate the general use of Multivariate Gaussian Mixture Modeling for selecting candidate genes for experimental characterization from synthetic lethality data sets. For the given example, integration of different data sources contributes to the identification of genetic interaction partners of arp1 and jnm1 that play a role in the same biological process.

A central goal of postgenomic research is to assign a function to every predicted gene. Because genes often cooperate in order to establish and regulate cellular events the examination of a gene has also included the search for at least a few interacting genes. This requires a strong hypothesis about possible interaction partners, which has often been derived from what was known about the gene or protein beforehand. Many times, though, this prior knowledge has either been completely lacking, biased towards favored concepts, or only partial due to the theoretically vast interaction space. With the advent of high-throughput technology and robotics in biological research, it has become possible to study gene function on a global scale, monitoring entire genomes and proteomes at once. These systematic approaches aim at considering all possible dependencies between genes or their products, thereby exploring the interaction space at a systems scale. This chapter provides an introduction to network analysis and illustrates the corresponding concepts on the basis of gene expression data. First, an overview of existing methods for the identification of co-regulated genes is given. Second, the issue of topology inference is discussed and as an example a specific inference method is presented. And lastly, the application of these techniques is demonstrated for the Arabidopsis thaliana isoprenoid pathway.

Background: Large scale screens for synthetic lethality are widely used in yeast genetics to systematically search for genes that are involved in specific biological processes. Often the amounts of data resulting from single screens far exceed the capacities of experimental characterization of every target found. Thus, computational tools are required to select promising candidates from a screen in order to reduce the number of experiments to