Behavioral Genetics and Animal Domestication - Genetics and the Behavior of Domestic Animals - page 51

architecture has been principally analyzed through QTL studies, whereby
wild populations are crossed with domestic breeds, with the resulting hybrids
(termed F1s) then either inter-crossed together again or back-crossed to one
of the parental populations. These individuals are then analyzed genetically,
allowing the major effect loci to be identified positionally on the genome,
and the relative strength of effect (i.e. how much they influence the phenotype)
estimated.
What all these diverse domesticated species appear to have in common is
that the domestication traits themselves all seem to be grouped together
in clusters or modules in the genome, rather than being randomly distributed
(i.e. a QTL for increased growth may lie near a QTL for decreased anxiety
behavior, rather than being very dispersed). Starting with domesticated plant
species, beans ( Koinange et al., 1996; P ´ rez-Vega et al., 2010 ), rice ( Cai and
Morishima, 2002 ), maize ( Doebley and Stec, 1991, 1993 ), and sunflowers
( Burke et al., 2002 ) all show this pattern. In animals such modules are also
once again seen when using similar wild domestic crosses. For example,

in the chicken various behavioral, morphological and life-history traits have
all been shown to be grouped together in clusters ( Karlsson et al., 2011a;
Sch ¨ tz et al., 2002; Wright et al., 2006b, 2008; 2010, 2012 ). Similarly,
a study using rat populations selected for high and low aggression also found
similar modules composed of QTL affecting physiological and behavioral
traits ( Albert et al., 2009 ). Silver foxes differentially selected for tame and
aggressive behavior have also been shown to exhibit correlations with other
dog-like characteristics (including skull shape and color) ( Hare et al., 2005;
Trut, 1999 ). In a QTL analysis of these animals, which was restricted to behav-
ior, overlaps between different behavioral measures were found ( Kukekova
et al., 2011 ). A cross between wild and laboratory zebrafish populations found
correlations between growth and boldness behavior in the inter-cross generation
( Wright et al., 2006a, c ).
When such modules do occur, a pertinent question is of course whether
each module represents one general “domestication-locus” (i.e. all the
domestication effects arise from a single pleiotropic locus), or whether they
are in fact comprised of many different, but closely physically linked, loci.
In the case of plants, there have been some examples where the actual causa-
tive genes and mutations have been identified, allowing this question to be
addressed more accurately. Where individual genes have been identified in
plants, these do indeed appear to have pleiotropic effects (pleiotropy here
refers to when a gene or QTL affects multiple different phenotypes). The
Q gene in wheat ( Faris et al., 2003; Simons et al., 2006 ) and the gene tb1
in maize ( Clark et al., 2004; Doebley and Stec, 1993 ) are classic examples.
However, with each of these examples, additional mutations and causative
genes have also been identified in separate, but physically linked, genes
( Doebley and Stec, 1991, 1993; Ji et al., 2006 ). In the animal examples it is
therefore tempting to suggest that these modules, similar to the pleiotropic

3