Mapping quantitative traits and strategies to find quantitative trait genes

Abstract

In 1999 a meeting took place at the Jackson Laboratory, a large mouse research centre in Bar Harbor, Maine, to consider the value of systematically collecting phenotypes on inbred strains of mice (Paigen and Eppig (2000) [1]). The group concluded that cataloguing the extensive phenotypic diversity present among laboratory mice, and in particular providing the research community with data from cohorts of animals, phenotyped according to standardized protocols, was essential if we were to take advantage of the possibilities of mouse genetics. Beginning with the collection of basic physiological, biochemical and behavioral data on nine commonly used inbred strains, the project has expanded so that by the beginning of 2010 data for 178 strains had been collected, with 105 phenotype projects yielding over 2000 different measurements (Bogue et al. (2007) [2].

Fig. 1

The distribution of three phenotypes in 16 inbred strains of mice shown as box and whisker plots. The y axis label is the name of the phenotype; the x axis labels are the strain names. Phenotypic data are from http://www.phenome.jax.org/pub-cgi/phenome/mpdcgi?rtn=docs/aboutmpd.

Fig. 2

A simulated example of the distribution of red blood cell counts in four consomic strains compared to 129S1/SvlmJ. The chromosomal structure of the consomic strains is indicated by their name so “chr 1” means that all the chromosomes are derived from 129S1/SvlmJ and chromosome 1 from the other progenitor, C3HeB/FeJ.

Fig. 3

The results of crossing a chromosome 1 consomic to its progenitor. Only four chromosomes are shown. The chromosomal structure is shown by colour, so that blue indicates descent from 129S1/SvlmJ and red from C3HeB/FeJ. Production of the inbred recombinant shown at the bottom of the figure is assumed to take a number of generations (i.e., it is not simply the product of intercrossing the F2 shown in the figure).

Fig. 4

The chromosomal structure of an F1, F2 and of recombinant inbreds. The chromosomal structure is shown by colour, so that blue indicates descent from 129S1/SvlmJ and red from C3HeB/FeJ. Below the recombinant inbreds are shown simulated results for their red blood cell counts (RBC).

Fig. 5

Results of mapping the using recombinant inbreds shown in Fig. 4 at six markers (data are given in Table 1). The x axis gives the position in arbitrary units, the vertical axis is the significance of the association, shown as a negative logarithm (base 10) of the P-value (log P).

Fig. 6

Mapping QTLs using an intercross: results on a single chromosome. The vertical axis shows the significance of the association expressed as a likelihood of odds or LOD score. In an intercross the LOD score is identical to a log P (negative logarithm (base 10) of the P-value) so the peak can be read as having a P-value of 10⁻²⁵. The dotted red lines indicate the extent of the 95% confidence interval.

Fig. 7

Simulation results for the confidence interval of a QTL, showing the effect of increasing the number of generations (x axis). The confidence interval is plotted on the y axis in centimorgans (cM). The effect of changing population size is also shown, with results for three breeding populations of size 50, 150 and 1500.

Fig. 8

The effect size (expressed as the proportion of variance attributable to the locus) for 843 QTLs mapped in an HS population of mice. Data are from .

Fig. 9

The relationship between heritability, plotted on the x axis, and the number of QTLs detected, plotted on the y axis. Data are from 843 QTLs mapped in an HS population of mice .

Fig. 10

Context dependent genetic effects. The top panel shows the effect of experimenter (plotted on the x axis) on a fear-related measure in mice, time in the open arms of an elevated plus maze, plotted on the y axis. The bottom panel shows the effect is dependent on genotype. Results are shown for a single marker with three genotypes (AA, AT and TT) shown on the x axis. The distribution of the phenotype associated with each genotype is shown separately for each experimenter.