Project 1: The Mouse Brain Library Project 2: Internet Microscopy (iScope) Project 3: Neurocartographer and Segmentation of the MBL Project 4: The Neurogenetics Tool Box


















Principal Investigator/Program Director Williams, Robert W.


Community mapping resources

The Genotyping and Mouse Colony Core describes a large advanced intercross that is already beginning to provide data to map and identify large numbers of loci that control variation in the architecture of the CNS. A major goal of the NTB is to provide Web-based software that will allow neuroscientists and geneticists, including those not currently familiar with QTL mapping, to collaborate with one another to exploit the resources of the MBL and NeuroCartography projects. The large size of this tenth-generation advanced intercross and its derivation by multiple generations of recombination will allow QTLs to be mapped with a precision of less than 4 cM. We expect that the discovery of genetic sources of structural variation will rapidly catalyze correlative analysis of functional and behavioral traits. The recombinant inbred strains mentioned in the next paragraph (Belknap 1998, Williams 1998) provide a superb resource for extensive brain-behavioral correlative genetic analysis.

Sets of recombinant inbred (RI) lines will provide another major shared resource to be exploited with the aid of the NTB. The Mouse Brain Library already has a very extensive collection of brains from RI strainsan average of more than 6 animals from each of 35 BXD strains and ~24 AXB/BXA strains. In the first few years of the grant we will add at least 30 RI strains (for example, the BXH and CXB strains are already in progress), for a cumulative set of well over 80 RI strains in the collection. RI strains provide multiple individuals for estimating trait values (Belknap 1998) . Using average trait value for 10 or 20 individuals greatly reduces the nongenetic variance, in effect, giving the trait a higher heritability. This, in turn, increases the likelihood of success in mapping QTLs (Darvasi and Soller 1997) . Using this approach, Toth and Williams (1999 a, b, c) recorded EEGs from an average of 12 animals from each CXB RI strain, and they succeeded in mapping at least one QTL controlling the sleep cycle (REM vs. slow-wave sleep) in micea trait that normally has a very low heritability (h2 < 0.3). Another key advantage of RI lines is that hundreds of different types of traits (behavioral, neuropharmacological, neuroanatomical) can be studied by many investigators over a period of years using one fixed set of genotypes. All data, whether genotypes or phenotypes, are cumulative. Using these far more extensive sets of data on entire strains of mice, investigators can detect genes and QTLs that have diverse or pleiotropic effects. The main limitation of RI strains is that their statistical power tends to be low due to the relatively small number of strains in each set. But this is not a major problem at this early stage of analysis, before any QTLs affecting CNS structure have been mapped. In fact, we would like to start with QTLs that have major effects because they may have more prominent behavioral consequences.



Next Topic