Molecular genetics of protein secretion in seeds

In this project we are using eQTL data to identify relevant gene clusters involved in different biological processes in Arabidopsis, yeast and C.elegans (protein folding for all three organisms, flowering and terpenoid biosynthesis for Arabidopsis). For all these organisms the expression of ~20000 genes was measured using microarrays for each member of populations of recombinant inbred lines (n=~80-200). Genetic markers of these RIL individuals will be correlated to expression of the genes to calculate LOD scores for each marker position. Subsequently, gene correlation networks (based on association between genes with similar genetic regulation) will be determined as function of the correlation threshold. The size of such dataset matrix will be ~28,000X200. Because the total number of pair-wise comparisons are (n-1)! were n is 28,000 genes, we will first use gene subsets with significant QTLs (LOD score > 2.5). Subsequently, the LOD score will be lowered, in 12 steps and for each threshold the total number of genes and networks will be scored. Similar type of analysis will be done using six different publicly available data inputs (three micro-array datasets from Arabidopsis, two from yeast and one from C.Elegans).