GRAPPLE - Iterative modelling of gene regulatory interactions underlying stress disease and ageing in C. elegans

Systems biology aims to model quantitatively how complex systems work, the resulting model being refined by means of an iterative loop of experimental testing and further modelling. Our project is unique in the sense that we will identify regulatory interactions underlying stress response and lifespan using a combination of natural genetic variation in gene transcription profiles, identification of regulatory and regulated genes and intensive computational modelling. The resulting networks will then be tested and refined using iterative perturbation experiments. In this way the new data will generate new, stronger network models to define genes for perturbation analysis, which in turn will furnish new data for modelling using new computational tools, and so on. The novelty and the strength of our approach is also based on simultaneous use of a variety of new methods for statistical analysis of networks, including those for network matching to be developed by the members of our team. To date, systems biology has not taken advantage of natural genetic variation in predicting the regulatory interactions underpinning important biomedical phenotypes. By combining the most powerful experimental and computational methods with the simplest model animal system, our project will significantly advance both our understanding of a complex regulatory system with direct relevance to human health and in the advancement of methodology that can be applied to other systems. Molecular and quantitative geneticists within this EU-wide group will work together with modellers to connect the dots from genome to phenotype, and on to predictive uses in biomedicine and healthcare.

We shall identify regulatory interactions underlying stress response and lifespan using a combination of natural genetic variation in gene transcription profiles, identification of regulatory and regulated genes and intensive computational modelling. The resulting networks will then be tested and refined using iterative perturbation experiments. The new data will generate new, stronger network models to define genes for perturbation analysis, which in turn will furnish new data for modeling using new computational tools, and so on. The novelty and the strength of our approach is also based on simultaneous use of a variety of new methods for statistical analysis of networks, including those for network matching to be developed by the members of our team. Partner 3 has built a collection of 200 recombinant inbred lines derived from a cross between the standard laboratory isolate of C. elegans (N2 Bristol) and a genetically and phenotypically divergent isolate from Hawaii (CB4856). These lines show extensive variation in stress tolerance and lifespan (see description to WP3) and provide a fantastic resource for eQTL analysis. Partner 1 comprises PIs who have led the study of stress-adaptive phenotypes rather than just response phenotype, as well as those who have developed databases of gene/longevity relationships and gene pathway and network modelling approaches. Partners 2 and 4 have led the field in constructing and analyzing network models and Partner 5 has developed novel network design techniques for comparative analysis of networks. Our project will combine the expertise of these 5 partners to build, analyze and iteratively test predictive models for the networks that underlie stress tolerance and lifespan.

Systems biology aims to model quantitatively how complex systems work, the resulting model being refined by means of an iterative loop of experimental testing and further modelling. Our project is unique in the sense that we will identify regulatory interactions underlying stress response and lifespan using a combination of natural genetic variation in gene transcription profiles, identification of regulatory and regulated genes and intensive computational modelling. The resulting networks will then be tested and refined using iterative perturbation experiments. In this way the new data will generate new, stronger network models to define genes for perturbation analysis, which in turn will furnish new data for modelling using new computational tools, and so on. The novelty and the strength of our approach is also based on simultaneous use of a variety of new methods for statistical analysis of networks, including those for network matching to be developed by the members of our team. To date, systems biology has not taken advantage of natural genetic variation in predicting the regulatory interactions underpinning important biomedical phenotypes. By combining the most powerful experimental and computation methods with the simplest model animal system, our project will significantly advance both our understanding of a complex regulatory system with direct relevance to human health and in the advancement of methodology that can be applied to other systems. Molecular and quantitative geneticists within this EU-wide group will work together with modellers to connect the dots from genome to phenotype, and on to predictive uses in biomedicine and healthcare. Timeliness Our approach, of using the simplest animal system, of iterative experimentation and model building, and of directly linking eQTLs to phenotypic outcomes provides an unrivalled opportunity to dissect a naturally variant regulatory network. In contrast to work in mammals, it is the ease and speed of iterative experimental testing and validation that makes our project particularly powerful in advancing the model describing regulatory interactions. This approach is leveraged by the availability of cheap whole genome sequencing, advanced models for network analysis and construction, and rapid methods for directed perturbation and validation experiments. The time is now ripe for the application of this approach to a complex system such as stress and longevity regulation in C. elegans. Members of our consortium are international leaders in the construction and integration (Nature Genetics 40: 181-8), biological (Nature 431: 308-12, Nature Genetics 36: 492-6) and statistical (Phys Rev E 72: 011903, 78: 020901) analysis of networks. Combined with our expertise in the stress response (PNAS 101: 16970-5, 103: 2977-8), high-throughput (Nature Genetics 38: 896-903), and quantitative eQTL analysis (PLoS Genetics 2: e22, 3:e34), this places us in a unique position to undertake this project as an integrated team and to drive forward this area of research.