220
L. S. Emery and J. M. Akey
comparisons looking for adaptively evolving loci during hominid evolution (Amato
et al. 2009). Many of these genes have been implicated in selection by multiple
studies; however, due to differences in the statistics and significance cutoffs used by
each study, the overlap between sets of candidate regions is far from perfect. For
instance, as of 2009, only ~14% of regions (722 regions) implicated in selection
were identified by more than one genome-wide scan (Akey 2009). Encouragingly,
many of the genes previously known to be targets of selection have also been
detected by genome-wide scanning methods. In particular, HBB, DARC, and LCT
have been detected by several different studies (Table 9.1). These regions serve
as a “positive control,” showing that genome-wide scans are identifying selected
regions with independent supporting evidence for selection. Additionally, scans
have identified numerous new possible targets of selection, some of which are
identified in multiple studies. For example, PCDH15 has been identified in at least
six different genome-wide scans. This gene is important for retinal and cochlear
function and, when mutated, can cause Usher syndrome type 1F (Ahmed et al.
2001) and autosomal recessive deafness 23 (Ahmed et al. 2003). PCDH15 has
also been associated with familial combined hyperlipidemia (Huertas-Vazquez et
al. 2010). EDAR encodes the ectodysplasin A receptor and has been identified
in 11 different genome-wide scans for selection. The EDAR protein mediates
ectoderm-mesoderm interactions during development, and mutations in EDAR can
prevent the normal formation of hair, sweat glands, and teeth in a disorder called
hypohidrotic ectodermal dysplasia (Bryk et al. 2008; Mou et al. 2008). Perhaps
the most compelling new signature of selection, identified by ten different studies,
is SLC24A5, which encodes sodium/potassium/calcium exchanger 5, a protein
required for proper melanogenesis. Although it was already shown that SLC24A5
variants contribute directly to human variation in skin pigmentation (Lamason et al.
2005), previous evidence for selection was minimal.
By examining the candidate selection genes identified by studies so far, we can
search for patterns in gene function. For example, gene ontology analyses of results
from individual scans have suggested that the targets of recent positive selection are
enriched for genes related to immunity, olfactory reception, metabolism, reproduction, pigmentation, and muscle development (Wang et al. 2006; Voight et al. 2006;
Williamson et al. 2007; López Herráez et al. 2009). Furthermore, when results from
multiple scans have been compiled, the best-supported candidate selection regions
have proved to be enriched for various metabolic processes, signaling, development,
and the cell cycle, among others (Akey 2009). Over and over again, the signatures
of selection have been found in genes related to metabolic processes and pathways
(López Herráez et al. 2009; Hancock et al. 2010b), which is consistent with previous
hypotheses regarding the recent evolution of human metabolism (Neel 1962; Babbitt
et al. 2011). One possibility is that humans have adopted a drastically different diet
from that of our hominin ancestors and have therefore adopted extensively. Such
recent metabolic evolution could be attributed to the development of agriculture and
corresponding new food sources, as well as the availability of novel food sources
in the environments encountered by humans during the peopling of the world
(Hancock and Rienzo 2008; Hancock et al. 2010b; Babbitt et al. 2011). Although the
L. S. Emery and J. M. Akey
comparisons looking for adaptively evolving loci during hominid evolution (Amato
et al. 2009). Many of these genes have been implicated in selection by multiple
studies; however, due to differences in the statistics and significance cutoffs used by
each study, the overlap between sets of candidate regions is far from perfect. For
instance, as of 2009, only ~14% of regions (722 regions) implicated in selection
were identified by more than one genome-wide scan (Akey 2009). Encouragingly,
many of the genes previously known to be targets of selection have also been
detected by genome-wide scanning methods. In particular, HBB, DARC, and LCT
have been detected by several different studies (Table 9.1). These regions serve
as a “positive control,” showing that genome-wide scans are identifying selected
regions with independent supporting evidence for selection. Additionally, scans
have identified numerous new possible targets of selection, some of which are
identified in multiple studies. For example, PCDH15 has been identified in at least
six different genome-wide scans. This gene is important for retinal and cochlear
function and, when mutated, can cause Usher syndrome type 1F (Ahmed et al.
2001) and autosomal recessive deafness 23 (Ahmed et al. 2003). PCDH15 has
also been associated with familial combined hyperlipidemia (Huertas-Vazquez et
al. 2010). EDAR encodes the ectodysplasin A receptor and has been identified
in 11 different genome-wide scans for selection. The EDAR protein mediates
ectoderm-mesoderm interactions during development, and mutations in EDAR can
prevent the normal formation of hair, sweat glands, and teeth in a disorder called
hypohidrotic ectodermal dysplasia (Bryk et al. 2008; Mou et al. 2008). Perhaps
the most compelling new signature of selection, identified by ten different studies,
is SLC24A5, which encodes sodium/potassium/calcium exchanger 5, a protein
required for proper melanogenesis. Although it was already shown that SLC24A5
variants contribute directly to human variation in skin pigmentation (Lamason et al.
2005), previous evidence for selection was minimal.
By examining the candidate selection genes identified by studies so far, we can
search for patterns in gene function. For example, gene ontology analyses of results
from individual scans have suggested that the targets of recent positive selection are
enriched for genes related to immunity, olfactory reception, metabolism, reproduction, pigmentation, and muscle development (Wang et al. 2006; Voight et al. 2006;
Williamson et al. 2007; López Herráez et al. 2009). Furthermore, when results from
multiple scans have been compiled, the best-supported candidate selection regions
have proved to be enriched for various metabolic processes, signaling, development,
and the cell cycle, among others (Akey 2009). Over and over again, the signatures
of selection have been found in genes related to metabolic processes and pathways
(López Herráez et al. 2009; Hancock et al. 2010b), which is consistent with previous
hypotheses regarding the recent evolution of human metabolism (Neel 1962; Babbitt
et al. 2011). One possibility is that humans have adopted a drastically different diet
from that of our hominin ancestors and have therefore adopted extensively. Such
recent metabolic evolution could be attributed to the development of agriculture and
corresponding new food sources, as well as the availability of novel food sources
in the environments encountered by humans during the peopling of the world
(Hancock and Rienzo 2008; Hancock et al. 2010b; Babbitt et al. 2011). Although the