Biology Reference
In-Depth Information
An ethological survey of these videotaped tests identified over 300 dis-
crete behavioral observations (traits) which could be reproducibly scored
from video records in a binary fashion, e.g., presence or absence. Examples,
each simply scored Yes or No, include: “Wagging Tail?”; “Stays at the front
wall of the cage?”; and “Ears pinned?” ( Kukekova et al., 2008 ). Evaluation
of these traits for informativeness, redundancy, reproducibility, and consis-
tency identified two overlapping sets of traits: (i) a minimal set of 50 traits
that reliably distinguished fox populations along a tame
aggressive axis
( Kukekova et al., 2008 ); and (ii) a larger, 98-trait set ( Kukekova et al.,
2011a ), that includes all traits from the minimal trait set plus additional traits
that capture other dimensions of fox behavior ( Kukekova et al., 2011a ).
Behahior of foxes from parental (tame and aggressive) and crossbred ped-
igrees was evaluated using this standardized testing protocol and videotaped.
The resulting videotape dataset comprised a total of 1003 foxes (83
tame foxes, 80 aggressive, 93 F1, 293 backcross-to-tame, 202 backcross-to-
aggressive, and 252 F2 foxes). Video records were analyzed to deconstruct
fox behavior by scoring for the presence or absence of each of 98 specific
binary traits ( Kukekova et al., 2011a ). Principal-components (PC) analysis
was then used ( Kukekova et al., 2011a ), to identify independent (i.e. uncor-
related) underlying factors (i.e. the principal components) that accounted for
decreasing amounts of the total variance in observed behavior ( Kukekova
et al., 2011a ). Specific methodological aspects of the PC analysis are
described in Kukekova et al., 2008 and 2011a.
In this large data set, the first two principal components, PC1 and PC2,
accounted for 33% and 9% of the total behavioral variation, respectively
( Kukekova et al., 2011a ). PC1 clearly distinguished tame foxes from aggres-
sive foxes; F1 foxes yield intermediate values that extend into the ranges of
both the tame and aggressive foxes. The scores of the backcross generations
resegregate ( Figure 10.2A ) such that mean values of PC1 in the different
populations defined a linear gradient of heritable behavior, ranging from
aggressive to tame, that clearly corresponds to the relative proportions of
aggressive to tame ancestry in each population ( Figure 10.2A ).
PC2 did not follow a similar gradient ( Figure 10.2B ). Review of the
discrete behavioral traits that contribute to these two principal components
demonstrated that PC1 and PC2 are comprised of very different aspects of
behavior ( Figure 10.3 ). PC1 is comprised of traits that distinguish overall
aggressiveness from tameness, whereas most of the traits important to PC2
can be interpreted as distinguishing bold from shy behavior (independent of
the degree to which a fox was tame or aggressive).
QTL MAPPING OF FOX BEHAVIOR
These two principal components of behavior were used as phenotypes to
identify associated quantitative trait loci (QTL) in informative fox pedigrees
Search WWH ::




Custom Search