[
International Worm Meeting,
2003]
The genome projects have paved the way for new genome-wide functional studies on several organisms. We can now use the rich supply of DNA microarray data to find expression patterns that reflect physiologically meaningful relationships. We have developed a computational technique for identifying conserved co-expression patterns in multiple organisms. Co-expression that represents a functionally-relevant coupling between two genes will confer a selective advantage to the organism and hence be selected over the course of evolution while irrelevant co-expression will be lost. We constructed a multi-species expression network combining microarray data from over 3000 microarray experiment from four evolutionarily distant organisms H. sapiens, S. cerevisiae, C. elegans, and D. melanogaster. We found over 10,000 examples of pairs of genes that are strongly co-expressed in microarray data from at least two organisms. These gene correlations form a network revealing many relationships between genes involved in ancient pathways and complexes, spanning several cellular and multicellular core processes. The network includes over 300 genes that encode novel proteins, and we can assign potential functions to these genes based on their co-expression with genes of known function. We are in the process of making the network available as a resource for the academic community.
[
International Worm Meeting,
2009]
Interactions between proteins are a key component of most or all biological processes. A key challenge in biology is to generate comprehensive and accurate maps (interactomes) of all possible protein interactions in an organism. This will require iterative rounds of interaction mapping using complementary technologies, as well as technological improvements to the approaches used. For example, we recently developed a novel yeast two-hybrid approach that adds a new level of detail to interaction maps by defining interaction domains(1). Currently, I am working to generate an interaction map of proteins involved in controlling cell polarity in C. elegans to improve our understanding of the molecular mechanisms that establish and maintain cell polarity in multicellular organisms. I will combine two fundamentally different interaction mapping techniques: the yeast two-hybrid system (Y2H) and affinity purification/mass spectrometry (AP/MS). This will provide more detail by identifying both direct interactions between pairs of proteins by Y2H, and the composition of protein complexes by AP/MS. Moreover, interactions missed by one technology may be detected by the other, leading to a more complete interaction map. I will integrate the physical interactions with phenotypic characterizations. To this end I will systematically characterize the interaction network in vivo using two distinct models of polarity: asymmetric division of the one-cell embryo, and stem-cell-like divisions of a multicellular epithelium (in collaboration with M. Wildwater and S. van den Heuvel). M. Boxem, Z. Maliga, N. Klitgord, N. Li, I. Lemmens, M. Mana, L. de Lichtervelde, J. D. Mul, D. van de Peut, M. Devos, N. Simonis, M. A. Yildirim, M. Cokol, H. L. Kao, A. S. de Smet, H. Wang, A. L. Schlaitz, T. Hao, S. Milstein, C. Fan, M. Tipsword, K. Drew, M. Galli, K. Rhrissorrakrai, D. Drechsel, D. Koller, F. P. Roth, L. M. Iakoucheva, A. K. Dunker, R. Bonneau, K. C. Gunsalus, D. E. Hill, F. Piano, J. Tavernier, S. van den Heuvel, A. A. Hyman, and M. Vidal, A protein domain-based interactome network for C. elegans early embryogenesis. Cell, 2008. 134(3): p. 534-545. .