Construction of 1000 recombinant inbred lines of genetically diverse mice for studying system genetics

 Dr. Fuad Iraqi

 Animal models of complex human disease are essential tools for understanding the genetic and functional basis of these conditions. Recombinant inbred lines (RIL) are populations of inbred lines descended from known founder strains and are very useful in the analysis of quantitative traits and gene expression in the mouse, because their genomes are fixed. However, at present the RILs available (such as the BXDs) are not optimal because they do not capture sufficient genetic variation, and the mapping resolution achievable is too low. Therefore it is important to derive a new set of RILs

descended from as genetically diverse set of founders as possible, and that can achieve
high mapping resolution.
The Collaborative Cross (CC) (Churchill et al. 2004: Iraqi et al. 2008) is a long-term international collaboration to construct a large panel of recombinant inbred lines (RIL) of mice and make them available to the scientific community, in a manner free of intellectual property constraints. Several panels of RIL mice already exist, such as the BXD, and are used extensively for mapping quantitative trait loci (QTL) and in genetical genomics experiments (Chesler et al. 2005). However, the existing RIL panels have severallimitations. First, the founder strains used to make the panels are closely related classical inbred strains (such as C57BL/6J and DBA/2Jin the case of the BXD) resulting in large segments of the genome that are identical by descent (IBD) and which therefore cannot harbour QTL. The number of SNPs segregating between wild-derived strains such as CAST/EiJ or WSB/J is much greater than among the classical strains Second, the number of lines in each panel (under 100) is too small. This means that the QTL mapping resolution is typically no better than an F2 cross (around 10cM), and therefore does not deliver single gene resolution (Valdar et al. 2005), and further that the

study of epistasis is hampered by the small number of independent genomes available. Nevertheless, the existing RIL have demonstrated the power of using a single replicable genetic reference panel to integrate data from many experiments, the principle advantage being that information (phenotypes, genotypes, gene expression etc) collected on different individuals from the same genetically identical line can be imported into a single database, and in some cases analysed as if collected on one individual; (Wang et al. 2003), http://www.genenetwork.org.

In order to find a way through these difficulties the Complex Trait Consortium (www.complextrait.org) proposed making a panel of approximately 1000 lines of mice descended form an eight-way cross of founder strains [A/J, C57BL/6J, 129S1/SvImJ, NOD/LtJ, NZO/HiLtJ, CAST/Ei, PWK/PhJ, and WSB/EiJ ], including wild-derived strains to increase genetic diversity (Churchill et al. 2004; Roberts et al. 2007). Recent analysis of the NIEHS/Perlegen mouse genotype data (Roberts et al. 2007) indicates that the choice of CC founders is close to optimal in the sense of capturing the maximum genetic diversity among inbred strains, because it includes three wild-derived strains (CAST, PWK and WSB). In short, far more QTL are expected to be segregating in the CC compared to an F2 intercross between two classical strains (where between 30% and 50% of the genome will be IBD), and QTL mapping accuracy will typically be under 1Mb using the CC compared to over 20Mb with an F2 (Valdar et al. 2005).

Once completed, these lines will form the stable genetic reference population for the dissection of complex traits in the mouse. A major use will be mapping quantitative trait loci (QTL). Simulations indicate that the CC will attain a mapping resolution of under 1Mb for a small-effect QTL and is also ideally suited for the analysis of phenomena such as epistasis (Valdar et al. 2005). Being RILs, it is possible to repeat phenotypic measurements on genetically identical individuals to reduce experimental variance or to vary environmental conditions to find gene by environment interactions.

Here, I propose to develop a 1000 RI CC mouse lines in Qatar research Institute which can lead to major gene mapping projects at the National and International levels.