International Mammalian Genome Society

The 13th International Mouse Genome Conference
October 31-November 3, 1999

Table of Contents * Structure * Bioinformatics * Sequence * Mapping * New Tools * Gene Discovery * Developmental * Mutagenesis * Functional Genomics

E7 Efficient Localization of Mutations by Interval Haplotype Analysis

David R. Beier and Isaac M. Neuhaus. Genetics Division, Brigham and Women's Hospital, Harvard Medical School, Boston, MA

Although the development of highly polymorphic genetic markers has profoundly enhanced our ability to perform linkage analysis, the execution of these analyses as single-point genome-wide screens does not exploit the considerable advantages of using inbred strains. In the case where these strains are defined with respect to allele sizes, the phase of the parental alleles is known and the genotypes of the progeny can be used to infer the haplotype of the interval between markers. This information can reduce the number of individual loci that need to be tested to comprehensively assess linkage. An efficient approach for linkage analysis is of particular utility given the number of mutations presently being generated using mutagenesis protocols.

We have described a strategy of haplotype analysis in which a predicted value for the number of chromosomes inherited as non-recombinant from an unaffected parent (based on the assumption of a Poisson distribution of recombination events) was compared to the experimentally determined value. This strategy has been revised such that no such assumption is made. This approach is simpler in that it is a comparison within a data set of the relative randomness of the inheritance of haplotype intervals; these intervals are defined by the genotypes of a single proximal and a single distal marker. We have developed a spreadsheet program for data analysis in which the relative randomness of haplotype inheritance can be compared by reporting the c2 inferred for an interval as a fraction of the maximum possible c2.

We tested this strategy in data sets simulating recessive inheritance of a mutation. Using an empirically-determined threshold for the identification of linkage, every interval carrying a simulated mutation was ascertained in 200 data sets, even when these contained as few as 5 intercross progeny. In small data sets the identification of falsely positive "linked" intervals is a modest problem (although it should be appreciated that even in these cases most of the genome can be readily excluded). However, an evaluation of this as a function of sample size reveals that with 20 or more meiotic events the number of incorrectly linked intervals is reduced to one or none.This simulation demonstrates that a genome-wide screen for the chromosomal localization of an autosomal recessive mutation can be performed specifically and sensitively using analysis of only 40 markers in 10 intercross progeny.

This strategy has been used to map 5 mutations: jck (juvenile cystic kidneys), wa3 (waved3), tpd (torpedo), wn (white nose), and wpk (Wistar rat polycystic kidney). Interval haplotype analysis has also been utilized for localization of QTLs affecting the severity of polycystic kidney disease by comparative analysis of cohorts at the extremes of a trait distribution. While it might be suspected that this strategy may be insufficiently sensitive to identify all QTLs with a high degree of confidence, this approach is a means to rapidly identify intervals containing loci with major effects.


Abstracts * Officers * Bylaws * Application Form * Meeting Calendar * Contact Information * Home * Resources * News and Views * Membership

Base url
Last modified: Saturday, November 3, 2012