-
[
International Worm Meeting,
2011]
RNAi screens have revolutionized genetic screens in the worm. Like any genetic screen, an RNAi screen relies completely on phenotyping accuracy. However, to date there have been no published genome-scale screens using automated quantitative phenotyping - all phenotyping has been manual and at best semi-quantitative. Furthermore, although over 50 distinct phenotypes have been examined at genome-scale, the single most important phenotype for any evolutionary studies - fitness - has been completely ignored. We present here three complementary quantitative methods to direct assess the effect of RNAi on fitness and provide strong evidence that these methods are more sensitive than any manual screening as well as yielding highly reproducible quantitative measures of phenotypic strength. We illustrate how we have applied these methods to the study of natural variation in C. elegans and demonstrate the critical importance of quantitation to identify subtle defects and to correct for strain to strain growth differences.
-
[
International C. elegans Meeting,
2001]
We know the complete genome sequence of C.elegans , and can therefore predict the sequences of nearly all encoded genes. A major problem remains to understand what roles each gene plays in the development and function of the worm. The approach that we are taking to this problem is to use RNA-mediated interference (RNAi) to inhibit the function of each gene in the genome. To do this, we are constructing libraries of dsRNA-expressing bacteria; each strain targets a single predicted gene by RNAi when fed to worms. This feeding technique, pioneered by Timmons and Fire, allows for high throughput RNAi screening and, furthermore, the libraries can be used for an unlimited number of future screens (for example, to identify genes with subtle or conditional phenotypes). We previously published the results of screening 87% of chromosome I genes (Fraser et al. 2000). We have now screened 88% of chromosome II genes and are currently screening an X chromosome library we have constructed. Data from chromosomes I and II are similar, with phenotypes detected for 14% of chr I and 12% of chr II genes. Also, as was seen on chromosome I, we find that fewer genes have an RNAi phenotype in the duplicated regions of chromosome II. We present an update of the project including a comparison of results obtained from chromosomes I, II, and X. In addition to identifying biological roles for many genes, we can also use our data to infer models of genome evolution and to discern relationships between the types of genes involved in different developmental processes. Finally, to increase our ability to identify RNAi phenotypes, we are screening for mutants with an increased sensitivity to RNAi; these should prove valuable for future RNAi-based screens.
-
[
West Coast Worm Meeting,
2002]
Although there is a phenomenal quantity of sequence data available, biological function has only been assigned to a small percentage of predicted genes in any metazoan. Understanding how genetic information relates to biological function at the level not only of a single gene but of an entire genome is thus a key problem in modern biology. One approach is to analyse loss-of-function phenotypes of every predicted gene in the genome -- this is the approach that we have taken in C. elegans. We have generated a reagent that uses RNA-mediated interference (RNAi) to individually inhibit ~90% of all ~19,000 predicted genes in the C. elegans genome. Using this reagent, we examined loss-of-function phenotypes for ~90% of all predicted genes and have found ~1700 genes to have detectable loss-of-function phenotypes. This is the first systematic functional analysis of a metazoan genome. We find striking differences in the functions encoded on different chromosomes, particularly the X chromosome, which has very few essential genes. In addition to differences between chromosomes, we find evidence for the clustering of genes of similar functions in multi-megabase regions of individual chromosomes. Genes in these regions tend to share transcriptional profiles, suggesting that the physical clustering we observe may facilitate transcriptional co-regulation. We also discuss the future prospects for the technique of genome-wide RNAi screening in the worm.
-
[
International Worm Meeting,
2003]
C. elegans is in many ways an ideal organism in which to examine signal transduction. The genome encodes components of almost all known pathways and these genes are organised in well-conserved modules. Furthermore, unlike many other metazoa, the output of signaling in the worm is mainly non-stochastic i.e. given a certain combination of signals in a particular cell, the developmental decision is always the same.In an ideal world, we would like to know all the components of all pathways and their ordering that is, the way that information flows and is processed in a pathway, and the ways that pathways overlap and thus integrate information. This information should ultimately lead to an explanation for signaling specificity and the amazing balance in most pathways between sensitivity to low signal input coupled to great robustness. We present a combined strategy to map out signaling pathways in the worm, including genome-wide RNAi screens, high-throughput yeast two-hybrid analysis and bioinformatic approaches. Each approach is independent and systematic, and combining them should allow relatively (!) unbiased assembly of signaling pathways. As well as these abstract fuzzy notions, we also present real unpublished data arising from firstly RNAi screens examining modulators of ras signaling in the vulva and secondly yeast two-hybrid screens mapping out a physical interaction space of signaling.
-
[
International Worm Meeting,
2005]
Complex genetic interactions underlie much of biology and are at the root of many human diseases. However, most frequently, our analysis in model organisms of the genetic basis for phenotypic traits involves the perturbation of single genes, rather than the systematic examination of gene interactions. We are attempting to explore genetic interactions more systematically using screens for synthetic lethals. A synthetic lethal interaction between two genes, A and B, occurs if deleting either A or B yields a viable organism, whereas removing both A and B is lethal. In yeast, systematic screens for pairs of synthetic lethal genes have uncovered a large amount about the wiring of basic cellular biology. We are identifying synthetic lethals in C. elegans by comparing the sets of lethal genes generated by RNAi in wild-type and in mutant backgrounds. To screen sufficient numbers of genes, we have developed an efficient high-throughput method for inducing RNAi by feeding and also for the automated microscopic analysis of the resulting phenotypes. We induce RNAi by feeding dsRNA-expressing bacteria to worms growing in liquid culture this can be done in 96-well format and is as efficient as other ways of RNAi by feeding. To analyse phenotypes, we have set up an automated image analysis system which we will describe this generates precise quantitative measurements of sterile and lethal phenotypes. These RNAi feeding and analysis protocols allow us to analyse ~1200 genes targeted by RNAi per day. We present the results of screens for genes that are synthetic lethal with
efl-1/EF2 gene; this gene is part of the
lin-35/Retinoblastoma complex that is involved in regulation of the cell cycle regulation and of vulval development. We find multiple novel biological connections including
ncl-1 and a component of SWI/SNF and with SynMuv genes. We will present these data and discuss our results.
-
[
European Worm Meeting,
2006]
. Wendy Wong and Andrew Fraser. C. elegans is a popular model system for the systematic analysis. However, relatively little is known for its networks of genetic interactions. We hereby construct an integrative interaction map for C. elegans by integrating diverse functional genomics Datasets. We will also present some of the novel tools in querying the interaction map for answering powerful biological hypotheses.
-
[
International Worm Meeting,
2009]
C. elegans (worm) is a powerful model organism for Systems Biology of multi-cellular metazoans, such as human, and recently genome-scale functional network for worm genes, WormNet, has been developed and was proved to be highly predictive loss-of-function phenotypes (1). Here, we present WormNet version 2, covering ~75% of the worm genes encoding proteins (15,038 worm genes) with about 1 million functional links between genes. To construct WormNet version 2, we integrated 21 heterogeneous genomics data sets (mRNA expression profiles by cDNA microarray, text-mining of scientific literature by co-citation, yeast 2 hybrid analyses, gene neighbors, phylogenetic profiles, genetic interactions, and conserved gene interactions transferred from yeast/fly/human). Various computational validations imply improved prediction ability for many loss-of-function phenotypes. Here, we used the network to predict novel genetic interactions among disease related genes in worm. From earlier experimental observation from large-scale genetic interaction screening (2), we found enrichment of genetic interactions between pathways. Assuming epistatic interactions between pathways, we predicted new genetic interactors to several disease genes. The new genetic interactors were predicted by guilt-by-association approach. Cross-validation analysis shows enriched connectivity among genes that genetically interact with each disease related gene. The experimental validations using RNAi experiments are under process and will be present in the meeting. Only about 5% of human diseases in developed country are monogenic. Therefore, understand of genetic interaction among disease related genes is critical in development of therapeutic methods for polygenic common diseases such as cancer and diabetes. Network-guided genetic interaction screening would allow construction of disease related pathway map with reduced cost and time in a model organism, C. elegans, and will shed light on understanding of human disease mechanisms in the future. Reference 1.Nature Genetics 40: 181 (2008) 2.Nature Genetics 38: 896 (2006).
-
[
International Worm Meeting,
2011]
Common human disorders are complex. Disease risk depends on multiple inherited genetic variations. Differences in genetic background between individuals thus have a major effect on the outcome of inheriting any single disease-related allele, and the main goal of this project is to begin to assess directly the effect of genetic background on loss of function phenotypes. We will use RNA interference in Caenorhabditis elegans to generate loss-of-function phenotypes in multiple different natural isolates. We can thus systematically examine how a given monogenetic perturbation leads to a diversity of phenotypic consequences in different individuals of the same species. Having identified genes with differing RNAi phenotypes between any two isolates, we then use standard mapping strategies to identify the quantitative trait loci accounting for the diversity. In this way we can examine the genetic basis for the variation in loss-of-function phenotype of each gene examined. This will give us a comprehensive view of how genetic background impacts phenotypic variation. We have almost completed this first for two isolates, N2 and Hawaii, and will present the results of our screening and analysis here. To facilitate this large-scale study, we are using a manual screen to score lethality, brood size defects, and growth defects. Using a novel quantitative assay, we can measure precisely the quantitative effect on fitness of RNAi against any gene and this is a powerful tool for accurate comparisons between isolates. This assay uses a standard worm sorter to count progeny over an extended period of 4 days and we will presenting our experimental design and preliminary findings.
-
[
European Worm Meeting,
2006]
Ben Lehner, Catriona Crombie, Julia Tischler, Angelo Fortunato and Andrew G. Fraser Most heritable traits, including disease susceptibility, are affected by the interactions between multiple genes. However, we still understand very little about how genes interact since only a minute fraction of possible genetic interactions have been explored experimentally. To begin to address this, we are using RNA interference to identify genetic interactions in C. elegans, focussing on genes in signalling pathways that are mutated in human diseases. We tested ~65,000 pairs of genes for possible interactions and identify ~350 genetic interactions. This is the first systematically constructed genetic interaction map for any animal. We successfully rediscover most components of previously known signalling pathways; furthermore, we verify 9 novel modulators of EGF signalling. Crucially, our dataset also provides the first insight into the global structure of animal genetic interaction maps. Most strikingly, we identify a class of highly connected ''hub'' genes: inactivation of these genes greatly enhances phenotypes resulting from mutations in many different pathways. These hub genes all encode chromatin regulators, and their activity as genetic hubs appears conserved across metazoans. We propose that these genes function as general buffers of genetic variation and that these hub genes will act as modifier genes in multiple, mechanistically unrelated genetic diseases in humans.
-
Na, Hong, Fraser, Andrew, Ray, Debashish, Hughes, Timothy, Ramani, Arun, Tan, June
[
International Worm Meeting,
2013]
Alternative splicing (AS) is a highly regulated process that contributes to proteome complexity. With recent advances in sequencing and other high-throughput technologies, our knowledge of the many AS events in C. elegans has rapidly increased. However, for many of these events, it is still not known which splicing factors are involved, and how they function to regulate each splicing decision. We have generated RNA-Seq, RNA binding, and genetic interaction data for several splicing factors to model how these factors regulate different AS events- whether individually, or combinatorially with other splicing factors.
We have identified AS events that are perturbed in several splicing factor-defective backgrounds- including loss-of-function of
asd-1,
fox-1,
mec-8,
sym-2 and
exc-7. Combining this data with RNA-binding specificities for these splicing factors, we have identified genes that are likely direct targets of these splicing factors, and are in the process of validating these splicing targets in vivo. In addition, we systematically probed for other factors that genetically interact with these splicing factors by screening an RNAi library targeting ~400 genes containing known or predicted RNA-binding domains. Scoring for enhanced population fitness defects, we have identified interactions with components of the spliceosome, regulatory splicing factors, as well as other factors involved in mRNA-processing. We are using this data as an added resource to predict factors that may function in the same splicing regulatory pathways, to identify instances of combinatorial splicing regulation.