Gene clustering based on RNAi phenotypes of ovary-enriched genes in C. elegans

Recently, a set of 766 genes that are enriched in the ovary as compared to the soma was identified by microarray analysis [1]. Here, we report a functional analysis of 98% of these genes by RNA interference (RNAi). Over half the genes tested showed at least one detectable phenotype, most commonly embryonic lethality, consistent with the expectation that ovary transcripts would be enriched for genes that are essential in basic cellular and developmental processes. We find that essential genes are more likely to be conserved and to be highly expressed in the ovary. We extend previous observations [2, 3] and find that fewer than the expected number of ovary-expressed essential genes are present on the X chromosome. We characterized early embryonic defects for 161 genes and used time-lapse microscopy to systematically describe the defects for each gene in terms of 47 RNAi-associated phenotypes. In this paper, we discuss the use of these data to group genes into "phenoclusters"; in the accompanying paper, we use these data as one component in the integration of different types of large-scale functional analyses [4]. We find that phenoclusters correlate well with sequence-based functional predictions and thus may be useful in predicting functions of uncharacterized genes.