GE52CH19_Wallace ARI 17 October 2018 10:54

Annual Review of Genetics

On the Road to Breeding

4.0: Unraveling the Good,
the Bad, and the Boring of
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

Crop Quantitative Genomics

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Jason G. Wallace,1 Eli Rodgers-Melnick,2

and Edward S. Buckler3,4
1
Department of Crop and Soil Sciences, The University of Georgia, Athens,
Georgia 30602, USA; email: [email protected]
2
Corteva Agriscience, DowDuPont, Johnston, Iowa 50131, USA
3
United States Department of Agriculture, Agricultural Research Service,
Ithaca, New York 14853, USA
4
Institute for Genomic Diversity, Cornell University, Ithaca, New York 14853, USA

Annu. Rev. Genet. 2018. 52:421–44 Keywords

First published as a Review in Advance on quantitative genetics, breeding, agriculture, adaptation, heterosis,
October 4, 2018
deleterious alleles
The Annual Review of Genetics is online at
genet.annualreviews.org Abstract
https://doi.org/10.1146/annurev-genet-120116- Understanding the quantitative genetics of crops has been and will continue
024846
to be central to maintaining and improving global food security. We outline
Copyright  c 2018 by Annual Reviews. four stages that plant breeding either has already achieved or will probably
All rights reserved
soon achieve. Top-of-the-line breeding programs are currently in Breeding
3.0, where inexpensive, genome-wide data coupled with powerful algorithms
allow us to start breeding on predicted instead of measured phenotypes. We
focus on three major questions that must be answered to move from current
Breeding 3.0 practices to Breeding 4.0: (a) How do we adapt crops to better fit
agricultural environments? (b) What is the nature of the diversity upon which
breeding can act? (c) How do we deal with deleterious variants? Answering
these questions and then translating them to actual gains for farmers will be a
significant part of achieving global food security in the twenty-first century.

421
 GE52CH19_Wallace ARI 17 October 2018 10:54

INTRODUCTION
The outlook for future food security is well known: By 2050 there will be approximately 9 billion
Quantitative people on the planet, and 9–11 billion by 2100 (50). The linear growth in food production seen
genetics: the study historically will not be enough to satisfy global demand by 2050 (142), especially with increasing
of traits that vary demand for high-quality protein by a growing global middle class (171). This argument is as old
continuously in a
as Malthus, yet to date agricultural production has managed to keep ahead of population growth
population instead of
having only a few thanks to mechanization, fertilization, plant breeding, and other agronomic advances. However,
discrete states it is dangerous to assume that just because production has outpaced population growth in the past
that it will inevitably do so in the future. Most of the easily arable land has already been brought
into cultivation, and owing to land degradation and urbanization, its area is actually shrinking
(198). Most major sources of water—both surface and groundwater—are being overdrawn, and
water shortages are likely to get worse in the next few decades (54). In addition, extreme weather
events, such as droughts, floods, and damaging storms, are predicted to increase as climate change
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

moves through the twenty-first century (17, 25, 27). In short, it is getting harder and harder to
grow crops, yet more and more people rely on us doing so.

THE FOUR STAGES OF PLANT BREEDING

Only 1–3% of modern industrialized society is directly involved in food production (187), a sharp
decline from rates in historical societies. Much of this shift is the result of management improve-
ments (e.g., plows, planters, fertilizers), but much is also due to improvements in crop genetics
through breeding (37). Although some important agricultural traits have discrete genetics—such
as Mendel’s famous pea phenotypes or major disease resistance loci—most are highly quantitative,
such as plant architecture, maturity, nutritional quality, and yield. Although quantitative genetics
is roughly a century old (43, 47, 188), the principles it encompasses have been applied throughout
the history of plant breeding.
Based on the techniques involved, we have split plant breeding into four major stages: three that
have already been achieved and one that we will likely reach in the near future (Figure 1). Each of
these stages builds upon earlier ones by integrating established techniques with new technologies
to increase breeding efficiency.
Breeding 1.0 began 10–12 thousand years ago, as people from around the world explored and
cultivated nearly 7,000 species of food plants (81). Professional breeders were rare or absent,
but phenotype-based selection by local, independent farmers eventually resulted in the dramatic
phenotypic changes seen in modern crops (36, 120).
Breeding 2.0 began in the late nineteenth and early twentieth centuries when inbreed-
ing depression was recognized (32), Mendelian genetics was (re)discovered (26, 34, 173), and
quantitative genetics theory was established (43, 188). Many advances in plant breeding dur-
ing this time were in the science of breeding itself, including replicated field trials, controlled
crossings, statistical analyses, formal experimental designs, hybrid breeding, pedigree-based es-
timates of breeding values, and precise measurement of yield at scale (e.g., with multirow
combines).
Roughly 30 years ago we entered Breeding 3.0 when molecular markers and genomic data
began to complement phenotypic data. This stage started with marker-assisted backcrossing and
pedigree confirmation, then moved to dissecting complex traits with linkage mapping (91). The
introduction of high-throughput genotyping then expanded the quantitative genetics tool kit to
dissect variation in natural populations (genome-wide association) (147) and to select on genome-
estimated breeding values (genomic selection) (119).

422 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

BREEDING 1.0
Incidental selection
by farmers

BREEDING 2.0
Statistical and
experimental design
to improve selection
effort
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

BREEDING 3.0
Integration of
genetic and
genomic data;
current state of
the art

BREEDING 4.0
Ability to combine
any known alleles
into optimal
combinations; will
be reached soon for
some crops

Figure 1
Four stages of plant breeding. Breeding efforts can be divided into four existing or near-future stages based on available methodology.
Breeding 1.0 is mostly incidental selection by farmers. Breeding 2.0 involves using statistics and experimental design to improve
selection efforts. Breeding 3.0 includes genetic and genomic data and is the current state of the art. Breeding 4.0 will probably arrive
soon (at least for some crops) and is marked by the genome-wide ability to combine any known alleles into desirable combinations.

We are now on the cusp of Breeding 4.0, a new level of breeding where functional genetic
variants can be rationally combined both faster and better than ever before. This level of breeding
is catalyzed by major technological advances in genetics and information systems. For example,
genome resequencing studies can now cost less than a replicated yield trial, and genome editing is

www.annualreviews.org • Crop Quantitative Genomics 423

 GE52CH19_Wallace ARI 17 October 2018 10:54

expected to enable parallel, precise modifications to many sites (possibly hundreds) per generation.
High-throughput phenotyping can measure numerous traits with unprecedented spatiotemporal
resolution, and machine learning approaches permit the processing and interpretation of agro-
Machine learning:
the use of computer nomic data at a level far beyond what humans can assimilate.
algorithms to extract Although one can conceive an even more advanced level (Breeding 5.0) involving de novo
patterns from large, design of genes, pathways, and traits, its realization is so far in the future that we will not spend
complex data sets with time on it here.
minimal human
guidance
Adaptive peak: OVERVIEW
the theoretical, ideal
genotype that is most In this review, we focus on several major questions facing quantitative genetics as it transitions
fit in a given from Breeding 3.0 to Breeding 4.0. Crop quantitative genetics has always straddled the boundary
environment between basic and applied science, so it is no surprise that these questions touch on basic mech-
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

Confounded: anisms of evolution, domestication, and development even as they impact plant improvement
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

a statistical practices and agronomic performance.

relationship where two We have chosen three key questions to consider: (a) How do we adapt crops to better fit
variables are correlated
agricultural environments? (b) What is the nature of the diversity upon which breeding can act?
so that their effects are
difficult or impossible (c) How do we deal with deleterious variants? We still have only partial answers to each of these
to separate questions, and fleshing them out is a prerequisite to fully harnessing the tools of Breeding 4.0.
Although many of our examples come from the grasses owing to the authors’ familiarity with
them, these issues apply across all crops.

HOW DO WE ADAPT CROPS TO BETTER FIT AGRICULTURAL

ENVIRONMENTS?
At the risk of stating the obvious, crops did not evolve in agricultural settings. Modern agriculture
puts plants in a very different environment from their ancestors, with many selective pressures
operating in different ways (Figure 2). As a result, new alleles are pushed to high frequencies, and
the resulting selective sweeps have been found in maize (72), wheat (18), rice (59, 190), soybean
(100, 202), tomato (84, 102), and sunflower (20), among others. Importantly, no crop has yet
reached the top of its adaptive peak, especially since changing agronomic practices keep moving
it. Finding ways to move crops closer to their peak is the ultimate goal of modern plant breeding.
We focus here on three aspects of agriculture where increased understanding is likely to benefit
breeding in the near term: new mega-environments, interplant competition, and soil interactions.

New Mega-Environments
A mega-environment is a group of environments sharing similar climatic conditions (day length,
rainfall, temperature, etc.). Most crops originated in very localized regions but were spread by
humans throughout the world (82). Adaptations to new mega-environments are thus among the
largest changes crops have gone through since domestication.
The fundamental problem with finding the genes involved in these adaptations, however, is
that population structure is usually highly confounded with the same adaptations one wants to
map (160). Although statistical methods can reduce false positives due to background population
structure (137), the correlation between adaptation and structure usually means they also reduce
true positives. Such confounding can be mitigated by decoupling population structure and adap-
tation, either through experimental design or by selecting populations where they are naturally
decoupled.

424 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

a Heterogeneous neighbors b Genetically similar/identical neighbors

Bet-hedging Human-focused
resource allocation resource allocation

Competitive
architecture

Cooperative
architecture
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

• Stable soil community • Disturbed soil community

• Unpredictable water/nutrients • Predictable water/nutrients

Figure 2
Plant adaptive context. (a) The environment in which crop wild ancestors evolved is very different from (b) the environment in modern
agriculture. Some of the changes involved are relatively well studied, such as repartitioning resources from vegetative tissue to grain and
fruit. Others, such as how plants evolve to live in a perpetually disturbed soil ecosystem, are only beginning to be understood.

In agriculture, the gold standard experimental designs to study adaptation usually follow either
a Multiparental Advanced Generation Intercross (MAGIC) (87) or Nested Association Mapping
(NAM) (117) design. Although the details differ, both methods separate adaptation from pop-
ulation structure by using a small number of parents to create a large number of recombinant
progeny. The resulting population has a balanced structure and retains the statistical power of
other controlled mating designs, allowing adaptive alleles to be identified with high confidence.
Such analyses have been used to identify genes controlling flowering time and photoperiod sensi-
tivity (14, 73, 107, 114, 155), drought tolerance (96), salinity tolerance (150), cold tolerance (152),
and development rate (124).
An alternate approach is to take advantage of populations where structure is naturally decou-
pled from selection. This situation usually occurs in organisms with large population sizes that live
across environmental gradients with few barriers to gene flow. Environmental genome-wide as-
sociation studies using such populations have proven extremely powerful and include plant height
and inflorescence architecture in sorghum (94, 122); photoperiod-, altitude-, and drought-related
adaptation in maize (4); climatic adaptation in conifers (195); a host of environmental conditions
in Arabidopsis (55, 93); and even salinity tolerance in Atlantic herring (112). Although not every
organism is suitable for this sort of analysis, taking advantage of those that are can identify genes
Photoperiod
and alleles that are adaptive for different environmental conditions. Breeders can then target these
sensitivity:
genes for improvement in related organisms. the tendency of many
plants to initiate
flowering based on
Interplant Competition how much light they
receive each day
The wild ancestors of crops rarely, if ever, existed in genetically homogenous stands, and even
traditional landraces are usually heterogeneous mixes of genotypes. Modern crops, however, often

www.annualreviews.org • Crop Quantitative Genomics 425

 GE52CH19_Wallace ARI 17 October 2018 10:54

grow in genetically uniform stands where the combined performance of a field is more important
than any individual plant. This imposes a form of group selection on modern cultivars, so that
plant architecture is optimized for whole-field production through steeper leaf and root angles,
Group selection:
a state where selection shorter plants, better tolerance of high planting densities, and similar traits (35, 37).
acts on the fitness of A competitive plant is one that is good at taking resources for itself at the expense of its neigh-
the entire group bors. Although several authors have noted that less-competitive plants yield better in modern
instead of individuals agricultural settings (e.g., 35, 127, 151), the genetic loci controlling competitiveness have only
rarely been mapped. This may be because the basic traits that underlie competition (root angle,
leaf angle, metabolite production) are usually interesting in their own right, so they are often
mapped without regard for competition (11, 16, 197). The few studies directly mapping com-
petition confirm that the beneficial loci in low-competition settings are different from ones in
high-competition settings (90, 166, 196). Especially at early breeding stages, both breeding and
research environments tend to reward plants that strongly compete with their genetically differ-
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

ent neighbors, even though these plants do not consistently deliver the best yields when planted
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

in uniform stands (41). Integrating selection for lower competition into most breeding pipelines
requires either alternative field schemes or (more likely) in silico modeling (203) to determine
optimal trait combinations to include in breeders’ selection indices.

Soil Interactions
Soil is not just a substrate for crop growth; it is arguably one of the most complex ecosystems on
Earth (30). Repeated tilling leaves the soil in a state of perpetual disturbance (both physically and
biologically) with highly altered microbial profiles that are further shifted by fertilizers, irrigation,
crop rotation, and other agronomic practices (33, 57, 58, 113, 128, 162, 184). Such disturbance
creates a situation where plants’ evolved responses are maladapted to the ecology of modern
agricultural soil. Nascent research seeks to understand the genetics of microbial associations,
such as in recruiting beneficial rhizobia (2, 29, 80, 164) and mycorrhizae (2, 95), establishing
microbial communities (3, 19, 174), excluding pathogens (5, 118), and potentially influencing
food quality (12). Microbial communities can alter host phenotypes (132, 179), although very
little is known about the quantitative genetics involved. Microbial profiling and metagenomic
analyses of environmental (170) and crop-associated (40, 49, 135) microbial communities could
open a window onto how to harness these communities for future breeding efforts. It should be
noted, however, that research in several plants indicates that the environment is probably the
biggest driver of plant microbial communities (77, 106, 135), especially in the rhizosphere. This
means it may be easier to adjust management practices or apply specific microbes directly to seeds
or plants than to breed for improved associations.

WHAT IS THE NATURE OF THE DIVERSITY UPON WHICH

BREEDING CAN ACT?
Although breeding can treat genetics like a black box (and did so in Breeding 1.0 and much
of 2.0), understanding the nature of plant genetic variation can significantly improve breeding
efforts. One of the best tools for this is the plant’s own genome because it contains a history of
environmental adaptation. Comparing different genomes within a species thus gives clues to which
loci are important; this sort of analysis is especially powerful in crops with hundreds to thousands
of genomes available (e.g., 1, 15).
A major conclusion of whole-genome comparisons is that plant variation is significantly driven
by variation in gene content and copy number instead of just differences in protein sequences.

426 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

This hypothesis was first put forward nearly 20 years ago (154) and suggested that the presence
and absence of genes might underlie important phenomena, such as hybrid vigor (46). Presence–
absence variation explains significant amounts of phenotypic variation (22, 104, 180), although
Presence–absence
many presence–absence genes have low RNA expression (68) and even fewer of them are translated variation: genetic
into proteins (182). It thus seems that, even though many genes can have variations in copy number, variation in the form of
not all such variations are important. Most of the important presence–absence variation in plants genes that are present
appears to stem from either polyploidy or tandem duplications, as opposed to other mechanisms in one individual but
completely missing in
like transposon duplication.
another
Polyploidy: the state
Polyploidy of an organism having
more than two
There have been at least two major polyploid events in all angiosperms, and many more in most complete genome
lineages (79, 186). Although this implies that polyploidy is the norm, most polyploid events are copies
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

actually evolutionary dead ends (115), and it is only the lucky few that survive in the long term. Tandem duplication:
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Even though plants readily undergo polyploidization, the fitness consequences of polyploidy when one or more
are still not completely understood (108). One known consequence is that many of the duplicated copies of a gene are
inserted next to the
genes are rapidly lost via mutation, deletion, or other mechanisms. This loss is not random; rather,
original
one syntenic segment (subgenome) from a given region tends to remain relatively intact, whereas
Transposon
the other is preferentially mutated, deleted, or otherwise degraded (156) (Figure 3). Both maize
duplication: the
and Brassica rapa, for example, show clear evidence of differential genome retention (21, 145). copying of a gene by
Bread wheat (Triticum aestivum) is an example of this process in action, as a substantial portion its hitchhiking with a
of its allelic variation stems from dosage-altering loss-of-function mutations stemming from its transposon that is
polyploidy approximately 10,000 years ago (88). Subgenome dominance is apparently established copying itself to a new
location
very rapidly, with both gene expression differences and epigenetic marks of dominance appearing
in the very first generation (39). Syntenic:
corresponding regions
in two genomes that
have similar gene
Tandem Duplication
structure due to
Tandem duplication of genes occurs because of replication errors or unequal crossing-over during sharing a common
meiosis (143). Since tandemly duplicated genes are often still under the control of their native reg- ancestor
ulatory elements, these duplications provide an easy mechanism for modification of gene dosage. Gene dosage: the
For example, variation in carotenoid degradation in maize is due to a dioxygenase gene that first number of copies of a
gene in an organism,
transposed approximately 2 Mb away and then was tandemly duplicated up to 23 times, providing
especially relative to
significant quantitative variation in carotenoid degradation (168). Tandem duplication has also other genes in the
been implicated in aluminum tolerance in maize (109), salt tolerance in wheat (200), and dwarfing same pathway
in both wheat (99) and sorghum (125). Convergent
evolution: the ability
of two organisms to
The Size of Mutational Space independently evolve
similar or identical
Given that both polyploidy and tandem duplication can lead to large increases in gene content (in-
traits
cluding significant variation within a species), how big is the actual mutational space in a crop? Or
in other words, how many different mutations can shift a crop toward the same goal? Although pop-
ulation genetics highlight the opportunities for convergent evolution (140), older studies suggest
the space for mutations can be remarkably small. For example, mutations in numerous genes can
produce the sweet corn phenotype, but three of four independent evolutions were all in one cleft
of a single enzyme (172). Meanwhile, rice aromas have developed via 10 independent mutations
in the same gene (86), and the nonshattering phenotypes of domestic sorghum, rice, and maize
are all due to independent mutations in the same orthologous genes (103). These were all recent

www.annualreviews.org • Crop Quantitative Genomics 427

 GE52CH19_Wallace ARI 17 October 2018 10:54

Duplicated genomes
A B A B A B

Random Subgenome
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

deletion dominance
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Figure 3
Subgenome dominance. Immediately after a polyploidy event, the genome contains two complete sets of
genes (A and B, center). Over time, many genes are reduced to a single copy owing to mutations and deletions
(red Xs). Although one would expect this process to be random (left), most genomes show evidence of genes
that were preferentially retained or lost in blocks (right). The dominant subgenome consists of the blocks
that are preferentially retained (blue boxes). Note that which subgenome a block belongs to depends only on
how well it is retained, not on where it came from. In other words, a species with both A and B genomes due
to polyploidy can have a dominant subgenome consisting of a mix of A and B segments, instead of A being
completely dominant to B or vice versa.

selections caused by humans, but natural variation follows the same pattern. For example,
sorghum and maize diverged about 12 million years ago, during which time maize underwent a
whole-genome duplication from which it currently retains 20–30% of the duplicated genes (156).
Despite this distance, large mapping projects in both species show consistent alignment of the
quantitative trait loci (QTLs), where one locus in sorghum is matched by two syntenic QTLs
in maize (107). All of this indicates that, in practice, there are only a limited number of routes to
alter a given phenotype.
What about raw genome size? Angiosperm genomes can vary over three orders of magnitude
(8), from the carnivorous herb Genlisea tuberosa (0.061 GB) (44) to the canopy plant Paris japonica
(149 GB) (136). At first glance, these differences in genome size can appear to have strong impacts
on QTL, especially ones that act through gene regulation. For example, in Arabidopsis (0.125–
0.150 GB genome), almost all QTLs are within 5 kb of the affected gene (185). By contrast, two
Quantitative trait
of the best-characterized QTLs in maize (∼2.6 GB genome)—teosinte branched 1 and vegetative
loci (QTLs):
locations in the to generative transition 1—are in enhancer elements approximately 60 kb away from the genes
genome that have been they affect (24, 153). However, even though the distance between important DNA elements can
identified as vary dramatically across plant genomes, the actual genomic space that is important seems to be
influencing a much smaller and more constant. For example, nearly 90% of maize phenotypic variation could
quantitative trait
be assigned to the 3% of the genome that is either protein-coding or noncoding open chromatin
(149). This gives a much smaller sequence space to search for variation. Similarly, only 5–7%

428 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

of the rice genome is in open chromatin regions that are likely to impact function (199). If this
pattern holds across other plants, it implies that the functional mutational space may be quite
small indeed.
Haplotype: a set of
alleles linked together
HOW DO WE DEAL WITH DELETERIOUS VARIANTS? on the same piece of
DNA
Human geneticists often focus on deleterious mutations because of their role in diseases, but most
plant breeders do not think about them much. Any obvious deleterious mutation is eliminated
early, with little thought given to molecular mechanisms or how many milder mutations go unseen.
Although Arabidopsis is estimated to accumulate 1 mutation per generation (131), maize appears
to accumulate nearly 90 of them (23). Assuming 5–10% of the genome has a functional role (149),
this means five to nine mutations per generation could potentially affect a phenotype, and most
of them would probably be detrimental. Identifying, controlling, and repairing these mutations is
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

likely to be a major aspect of research as we move toward Breeding 4.0.

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Deleterious Alleles and Domestication

When researchers compare domesticated species with their wild relatives, they find that domesti-
cation is often associated with an increase in the number of apparently harmful alleles. This pattern
can be seen in both plants and animals, including rice (105, 126), sunflower (144), tomato (84),
dogs (110), and horses (159). On the basis of population genetic theory, there are a few processes
that may explain this increase in detrimental alleles. First, domestication alters selection pressures
so that traits favored in the wild become neutral or disfavored under domestication (74). Second,
purifying selection is frequently reduced after domestication bottlenecks (83). Third, inbreeding
of domesticated varieties further reduces their effective population size and effective recombina-
tion rates (105). Although the relative contribution of each force varies by species, all of them
probably act to some extent on domesticated crops.
Emerging evidence suggests that this cost of domestication can be tempered by modern im-
provement practices. For example, modern inbred lines of maize carry fewer nonsense mutations
than their wild relatives (22) and have less genetic load than traditional outcrossing landraces
(193). Young alleles generally appear to be under more stringent purifying selection in maize
than in its ancestor, teosinte (7). In cassava, meanwhile, domestication loci have fewer deleterious
mutations in cultivated varieties compared to progenitor populations even as drift has increased
plants’ overall genetic load (141).

The Problem with Chromosomes

One problem with deleterious variants is that the biology of chromosomes makes them hard to
breed out. Organizing genes into large linear chromosomes helps cells properly segregate complete
genomes during division, but this structural constraint also reduces the efficiency of selection by
linking alleles that may have different (and often opposite) fitness values. This means that any given
individual has a mix of good and bad alleles linked together within a haplotype, and recombination
is often not efficient enough to ever create a single ideal combination. Instead, multiple suboptimal
haplotypes selectively interfere with one another so that none reach fixation. This phenomenon is
called Hill-Robertson interference (65) and results in negative linkage disequilibrium (repulsion
phase) between beneficial variants in low-recombination regions (42).
Hill-Robertson interference has important (but often overlooked) consequences for plant
breeding. First, interference reduces the efficacy of selection on any individual site (65). This

www.annualreviews.org • Crop Quantitative Genomics 429

 GE52CH19_Wallace ARI 17 October 2018 10:54

allows slightly deleterious variants to accumulate and reduces the probability of fixation for any
given beneficial allele, a process that has been shown in rice (105), maize (148), and sunflower
(144). Second, having beneficial alleles spread across multiple nonrecombining haplotypes reduces
Heterosis: the
tendency of a hybrid genetic variance by eliminating extreme phenotypes, which in turn decreases the raw material for
offspring to be more fit selection to act upon. Third, assuming most deleterious alleles are at least partly recessive, the
than either of its low-recombination regions where interference is most severe should benefit more from heterozy-
inbred parents gosity than high-recombination regions. This benefit arises because low-recombination regions
have a greater chance to complement deleterious alleles. Artificial inbreeding should thus favor
individuals that retain heterozygosity in low-recombination regions, as has been verified in oat (71)
and maize (117). Interference and its consequences are easiest to see around centromeres because
they are so large, but the same processes also occur in localized regions across the genome.
One can assume that at least some QTLs of agronomic importance are located in similar
low-recombination regions, making their improvement extremely difficult. One work-around is
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

to select for haplotypes that complement each other in the hybrid state, which is what results
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

in distinct heterotic groups in maize and other hybrid breeding programs (169). Looking toward
Breeding 4.0, technologies that allow breeders to precisely manipulate recombination sites or alter
specific alleles through genome editing could bypass the Hill-Robertson effect entirely.

Implications for Heterosis

Heterosis has been known for 150 years (31), but its molecular underpinnings are still debated
(157). Because of the complementation mentioned earlier, regions with strong Hill-Robertson
interference are expected to behave as a single overdominant locus (98) even though the individual
alleles are strictly dominant. This effect is called pseudo-overdominance.
Experimental work confirms that at least some apparently overdominant loci are actually
pseudo-overdominant. Several generations of random mating in a biparental maize family elim-
inate the appearance of overdominance (48, 121), and grain QTLs consistently localize to the
centromere and pericentromere (92, 158), which are both predicted to show strong pseudo-
overdominance. Confirming this prediction requires fine-mapping traits, something that by def-
inition is difficult to do in low-recombination regions but has been managed occasionally. For
example, a single overdominant QTL near the centromere of maize chromosome 5 could be
separated into two dominant QTLs in repulsion phase (53). Meanwhile, a sorghum height QTL
separated into two genes with opposite effects approximately 3 Mb apart (98); both alleles were
dominant and resulted in apparent overdominance in the hybrid.
Although we have focused on low-recombination regions, these regions usually have rela-
tively few genes in them (6, 38, 75). Therefore, even if the most obvious genetic load is in low-
recombination regions, the most overall load is likely in high-recombination regions. Selection is
more efficient in these regions, but other factors associated with high recombination can oppose
movement up the fitness landscape. Most notably, GC-biased gene conversion likely maintains
some deleterious mutations at frequencies much higher than expected under mutation-selection-
drift equilibrium (52, 149).
The importance of low-recombination regions for heterosis may also depend on the species.
In bread wheat, for example, heterosis appears to be more driven by epistatic effects than by
dominance effects (78). Significant heterosis has also been seen from gene-dosage effects in rice
(70), maize (194), and Arabidopsis (45), indicating that the relative amounts of different genes
are just as important as dominant and recessive relationships. Heterosis also changes in different
environments (101), which makes it more difficult to pin down consistent mechanisms.

430 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

DELIVERING QUANTITATIVE GENETICS TO THE FIELD

How important is it to understand the adaptive and deleterious variants across the genome? This
sort of knowledge has already proven useful in a few specific pathways, especially for traits that
are difficult and/or expensive to score and that are controlled by a small number of genes. Such
traits include many disease resistance loci, submergence tolerance in rice (189), and carotenoid
accumulation in maize (56, 192). For Breeding 3.0 and 4.0 to achieve their full potential, we must
identify how to translate basic knowledge into real-world results.

Fisher-Orr and the Diminishing Returns of Breeding

The basic unit of modern crop breeding is the QTL. The more beneficial QTL you can breed
into a variety, the better that variety becomes and the higher up the adaptive peak it climbs. First-
generation QTL mapping focused on the big domestication loci (e.g., 13, 85, 103, 139, 165, 167,
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

183) and disease resistance genes (e.g., 69, 111, 163, 191). These efforts were very successful, but
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

as time passed, the pattern emerged that newer QTLs generally had smaller effect sizes, especially
in outcrossing species. Smaller QTLs require more effort to map and provide less benefit, putting
crop breeding in a state of diminishing returns.
Several phenomena contribute to this pattern. First, scientists were rational: They cloned the
biggest QTL first. But another major factor was that most QTL effects appear to follow the
Fisher-Orr geometric model (129, 130). In brief, this model says (a) large-effect mutations are
more likely to be harmful than helpful and (b) selection is less efficient for complex traits than
for simple ones. In breeding contexts, this means that alleles with large beneficial effects rapidly
become fixed, after which only successively smaller and smaller allele effects actually move plants
closer to the adaptive peak (Figure 4). In addition, most beneficial alleles for complex traits (e.g.,
yield) have small effects in the first place.
Large mapping panels have confirmed that most alleles have small effects (14, 116, 133, 180).
This implies that, especially in major crops, many of the large-effect QTLs have probably already
been identified and fixed in elite germplasm, leaving breeders to work with progressively smaller
effects that are harder to identify. This appears to be especially true in outcrossing species like
maize (14, 134) and even cattle (reviewed in 64). Self-pollinating species appear to have more large-
effect genes, such as the dwarfing genes of the Green Revolution (60), flowering time in barley
(114), and drought tolerance in rice (178). It may be that in outcrossing species, the constant
reshuffling of haplotypes each generation selects for small-effect alleles that “play nice” with other
genes. By contrast, self-pollinating crops pass down complete haplotypes to their offspring, which
may more readily allow large-effect alleles to evolve.

Genomic Prediction
One method of dealing with increasingly small QTLs is to stop trying to map them and instead
work with the entire genome. This is the approach of genome-wide prediction and genomic
selection (Figure 5), a pair of incredibly successful paradigms that are affecting almost all sectors
of breeding (63). Genome-wide prediction is the process of using genetic data to predict an
individual’s phenotype; genomic selection is simply using those predictions to make breeding
decisions. This sort of selection scheme was first demonstrated by Meuwissen et al. (119), and its
goal is to improve breeding by both reducing the cost of phenotyping and reducing the cycle time
for early generation selections.
Genomic selection has proven extremely beneficial to dairy cattle, and signs indicate it will be
similarly valuable for crop breeding (28, 63, 176). However, there are still many unknowns, such

www.annualreviews.org • Crop Quantitative Genomics 431

 GE52CH19_Wallace ARI 17 October 2018 10:54

Adaptive
peak

Maximum
size of

Magnitude of trait change

beneficial
mutations
Beneficial

Harmful
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Beneficial

Harmful

Figure 4
The Fisher-Orr model. The Fisher-Orr model states that the closer an organism is to an adaptive peak, the more likely a large mutation
is to be harmful. The blue individual is far from the peak, so many different mutations result in a net benefit. By contrast, the green
individual is close enough to the peak that even moderately sized changes can overshoot the ideal and leave its fitness lower than before.

as the optimal allocation of resources or choice of experimental designs. Krchov & Bernardo
(89), for example, found that some breeding designs were always better under genomic selection,
some were always better under phenotypic selection, and many could shift between the two based
on budget, resource allocation, trait heritability, and other factors. Heffner et al. (61) estimate
that genomic selection outperforms marker-assisted selection in terms of gain per unit time as
long as the trait has a heritability of 0.2 or greater. Meanwhile, the choices of statistical models
for genomic selection are many and growing (51), although with real-world data there seem to
be few practical differences among them (62). Genomic selection for hybrid performance is also
becoming more powerful (76, 97), which is likely to make genomic selection of hybrids possible
even for historically inbred crops like wheat (201).
Genomic selection has great potential for improving crops, but in most cases that potential is
still unrealized. This is most true for species outside of the major row crops and outside of the
developed world, largely due to simple logistics. Genomic prediction requires high-throughput
genotyping systems and bioinformatics expertise that many programs do not have, and breeding
decisions for annual crops must occur in two to three brief windows for genomic selection to be
worthwhile. The cost of genotyping can also be a major determinant of whether genomic selection
is worthwhile (89, 146), though the still-falling price of sequencing implies that may not be true
for long. Prediction models are also limited to highly related germplasm; models using different
breeding programs rarely have value, and accuracies can rapidly drop even beyond half-sib family
structures (10). All of these barriers need to be overcome for genomic selection to see truly global
deployment.

432 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

a Traditional selection b Genomic selection

yi = μ +
∑mzijujδj + ei
Phenotype Training cycle (Re)train
(slower, expensive) model

Phenotype
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Genotype Selection cycle Predict via

model
(faster, less expensive)

Make crosses

Make crosses

Figure 5
Genomic selection. (a) Traditional selection involves cycles of making crosses to get new genetic combinations, evaluating the new
varieties by phenotyping, and using those evaluations to select the next generation of parents. (The point at which a variety is spun off
into production is not shown.) (b) Genomic selection adds several layers to this scheme and splits the process into training and selection
cycles. When training, the breeder must go around the entire outside loop: making crosses, getting genotype and phenotype
information from them, building a mathematical model, and finally using that model to select future parents. After that, the breeder can
skip the training portion and go directly from new material to genotype to choose parents based on the model, with no need to
phenotype. Although the model must be retrained every few generations, breeders generally save significant time and money by
skipping the (usually laborious and expensive) phenotyping step two or three times.

Democratizing Breeding 3.0

Many crop breeding programs are still at the Breeding 1.0 or 2.0 stages, especially in the developing
world, yet widely deploying Breeding 3.0 is critical to providing benefits for the global population.
This democratization process is already underway, as inexpensive genotyping catapults formerly
neglected crops into twenty-first century genetics. Millets (9, 175, 177, 181), cassava (138),
cacao (123), strawberry (67, 161), sweet potato (66), and many others are benefiting from this
technology, but it must reach even further to have truly global benefits. DNA sequencing is likely
to drop to $1/Gb in the next several years, which implies that virtually any breeding line can be
skim-sequenced for little cost. High-throughput sequencing facilities, meanwhile, will probably
reduce the price of genotyping seeds to only $1 or $2 per sample. However, while sequencing and

www.annualreviews.org • Crop Quantitative Genomics 433

 GE52CH19_Wallace ARI 17 October 2018 10:54

genotyping are dropping to negligible prices, field work and the logistics of tracking and processing
thousands of samples are not. New ideas and informatics systems are needed to enable small pro-
grams to deal with these logistics. Most breeders do not have the time or skills to process the data
themselves, and ideally breeders should never even need to look at a nucleotide sequence. Projects
like the Genomic Open-source Breeding Informatics Initiative are starting this effort, but a larger
international community of both public and private entities is needed to make Breeding 3.0 a global
reality.

MOVING TO BREEDING 4.0

Beyond just democratizing Breeding 3.0, what do we need to move to Breeding 4.0 and capitalize
on the ability to manipulate individual genes and even single base pairs? The last decade has seen
tremendous strides for identifying functional variants through genome-wide associations, evolu-
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

tionary constraint, comparative genomics, and a range of molecular biology assays. For breeding,
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

however, each of these approaches needs to be carefully measured in terms of its usefulness. Unlike
human genetics, crops do not have one focus species but rather hundreds of species, each with
many different breeding targets around the globe. A major question for the next decade is how to
leverage knowledge across these species and breeding programs to make the most use of limited
resources. For example, how can maize breeding learn from rice trials, what lessons can cassava
take from apple or cotton breeding, and how much of what is learned in Arabidopsis can we apply
to bananas, blueberries, oats, or loblolly pine?
A detailed understanding of every gene’s function is probably not needed for Breeding 4.0. We
do, however, need to be able to estimate the effect of each functional variant in a range of target
environments. On a practical level, how do we get there? Although a species could have 50 million
common variable sites in the genome, most of them probably do not matter very much. To reach
Breeding 4.0, we need to reduce this sea of variation to a few tens of thousands of high-probability
functional sites.
What types of data are the most useful for this filtering? We present some recommendations
below:
1. Genetic mapping of complex target traits is by far the most expensive but also the most
important approach to identifying functional variants. Complex traits such as yield rarely
resolve to single genes, and trying to resolve any given QTL to a single nucleotide is rarely
worth the cost. Instead, the central goal of mapping should be to resolve a few key QTLs to
the gene level to highlight pathways and processes not previously considered. However, the
greatest value of genome-wide association studies and genetic mapping may well be the data
sets of genotypic and phenotypic data they produce. These can be used to benchmark every
other approach, especially in the context of multiple environments, and can be integrated
into ongoing breeding efforts.
2. Genomic annotation is the process of distilling our molecular understanding of the genome
down to machine-processable data. It includes annotations with protein domains, gene on-
tologies, methylation patterns, chromatin states, transcription factor binding sites, expres-
sion levels, and many others. Although all of these measures may be biologically interesting,
there are very strong correlations among them, and many can be predicted from each other.
The community needs to rigorously define the cost-effectiveness of each annotation to
identify where to best put limited resources.
3. The mapping of intermediate phenotypes—for example, RNA transcripts or metabolite
abundance—gives much higher resolution than the mapping of terminal phenotypes such

434 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

as yield and frequently identifies the exact gene involved or even specific causal variants.
Related approaches such as chromatin profiling and proteomics could dramatically improve
our ability to identify functional variants. The key challenge is to make these technologies
Linkage drag:
cheap enough that crops can be profiled across a range of genotypes and environments. the process whereby
4. Evolution is the ultimate yield trial, as it integrates the success of various genotypes over introgressing a
millions of years. The greatest limitation of current approaches is that sequencing distant desirable allele into
evolutionary species only provides information on the most conserved elements. Saturation breeding lines brings
along many
of closely related species is needed to fully understand regulatory conservation. Although undesirable alleles
crops are grown in environments where they did not evolve, the careful choice of landraces linked to it
or wild analogs (as opposed to elite lines) may still be able to address these questions.

Each of these approaches can enrich for functional variants but not identify them for certain.
Intelligent integration of these approaches, however, could provide whole sets of high-confidence
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

variants in a cost-effective manner. This integration requires characterizing global germplasm

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

resources both genetically and phenotypically, developing informatics tools to share this informa-
tion, and using appropriate methods (e.g., machine learning or similar) to integrate these diverse
data sets.
A final key to Breeding 4.0 is large-scale genomic editing (tens to hundreds of sites per gener-
ation). Direct genome editing will then almost certainly replace crosses as the most efficient way
to tailor genetic variation into optimal combinations. Better still, it can do so without linkage drag
destabilizing the decades of breeding that went before. Of course, such edited crops would prob-
ably need to overcome consumer resistance to engineered foods, but that is an entirely different
aspect to global food security.
Breeding has always been a numbers game, so we do not need to identify every variant with
100% accuracy. Even if we are only right 10% of the time, it may be enough to push crop breeding
faster and more cost-effectively than we could otherwise. The hunt for these variants is already
underway in many labs around the world. Integrating all this work into breeding pipelines will
be key to providing an adequate, nutritious, and sustainable food supply for the entire globe
throughout the twenty-first century.

SUMMARY POINTS
1. Breeding can be divided into four stages based on available technology; we are currently
in Breeding 3.0.
2. To reach Breeding 4.0, we need to identify specific alleles that are responsible for desirable
variation in crops.
3. Crops are still only partly adapted to agriculture and can be improved by breeding them
to better fit modern growing environments.
4. Variation that is relevant to agriculture is not randomly spread through the genome, so
finding the relevant portions can improve the focus of breeding efforts.
5. Much of Breeding 4.0 probably revolves around identifying and purging deleterious
variants.
6. More effort is needed to democratize current and future breeding technologies so that
they benefit all of global agriculture.

www.annualreviews.org • Crop Quantitative Genomics 435

 GE52CH19_Wallace ARI 17 October 2018 10:54

DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS
We wish to thank Sara Miller for help preparing the manuscript, and the United States Department
of Agriculture Agricultural Research Service, the University of Georgia, and DowDuPont for
funding support that made the preparation of this article possible.

LITERATURE CITED
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

1. 3,000 Rice Genomes Project. 2014. The 3,000 rice genomes project. Gigascience 3:7
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

2. Afkhami ME, Stinchcombe JR. 2016. Multiple mutualist effects on genomewide expression in the tripar-
tite association between Medicago truncatula, nitrogen-fixing bacteria and mycorrhizal fungi. Mol. Ecol.
25(19):4946–62
3. Agler MT, Ruhe J, Kroll S, Morhenn C, Kim S-T, et al. 2016. Microbial hub taxa link host and abiotic
factors to plant microbiome variation. PLOS Biol. 14(1):e1002352
4. Romero Navarro JA, Wilcox M, Burgueño J, Romay C, Swarts K, et al. 2017. A study of allelic diversity
underlying flowering-time adaptation in maize landraces. Nat. Genet. 49:476–80
5. Álvarez-Pérez JM, González-Garcı́a S, Cobos R, Olego MÁ, Ibañez A, et al. 2017. Use of endophytic
and rhizosphere actinobacteria from grapevine plants to reduce nursery fungal graft infections that lead
to young grapevine decline. Appl. Environ. Microbiol. 83(24):e01564-17
6. Anderson LK, Lai A, Stack SM, Rizzon C, Gaut BS. 2006. Uneven distribution of expressed sequence
tag loci on maize pachytene chromosomes. Genome Res. 16(1):115–22
7. Beissinger TM, Wang L, Crosby K, Durvasula A, Hufford MB, Ross-Ibarra J. 2016. Recent demography
drives changes in linked selection across the maize genome. Nat. Plants 2:16084
8. Bennett MD, Leitch IJ. 2012. Angiosperm DNA C-values database (release 8.0, Dec. 2012). Kew R. Bot.
Gard. http://data.kew.org/cvalues/
9. Bennetzen JL, Schmutz J, Wang H, Percifield R, Hawkins J, et al. 2012. Reference genome sequence of
the model plant Setaria. Nat. Biotechnol. 30(6):555–61
10. Beyene Y, Semagn K, Mugo S, Tarekegne A, Babu R, et al. 2015. Genetic gains in grain yield through
genomic selection in eight bi-parental maize populations under drought stress. Crop. Sci. 55:154–63
11. Biscarini F, Cozzi P, Casella L, Riccardi P, Vattari A, et al. 2016. Genome-wide association study for
traits related to plant and grain morphology, and root architecture in temperate rice accessions. PLOS
ONE 11(5):e0155425
12. Bokulich NA, Thorngate JH, Richardson PM, Mills DA. 2014. Microbial biogeography of wine grapes
is conditioned by cultivar, vintage, and climate. PNAS 111(1):E139–48
13. Brewer MT, Moyseenko JB, Monforte AJ, van der Knaap E. 2007. Morphological variation in tomato:
a comprehensive study of quantitative trait loci controlling fruit shape and development. J. Exp. Bot.
58(6):1339–49
14. Buckler ES, Holland JB, Bradbury PJ, Acharya CB, Brown PJ, et al. 2009. The genetic architecture of
maize flowering time. Science 325(5941):714–18
15. Bukowski R, Guo X, Lu Y, Zou C, He B, et al. 2015. Construction of the third generation Zea mays
haplotype map. GigaScience 7(4):gix134
16. Burridge J, Jochua CN, Bucksch A, Lynch JP. 2016. Legume shovelomics: high-throughput pheno-
typing of common bean (Phaseolus vulgaris L.) and cowpea (Vigna unguiculata subsp, unguiculata) root
architecture in the field. Field Crops. Res. 192:21–32
17. Cai W, Borlace S, Lengaigne M, van Rensch P, Collins M, et al. 2014. Increasing frequency of extreme
El Niño events due to greenhouse warming. Nat. Clim. Chang. 4(2):111–16

436 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

18. Cavanagh CR, Chao S, Wang S, Huang BE, Stephen S, et al. 2013. Genome-wide comparative diversity
uncovers multiple targets of selection for improvement in hexaploid wheat landraces and cultivars. PNAS
110(20):8057–62
19. Chaparro JM, Badri DV, Vivanco JM. 2014. Rhizosphere microbiome assemblage is affected by plant
development. ISME J. 8(4):790–803
20. Chapman MA, Pashley CH, Wenzler J, Hvala J, Tang S, et al. 2008. A genomic scan for selection reveals
candidates for genes involved in the evolution of cultivated sunflower (Helianthus annuus). Plant Cell
20(11):2931–45
21. Cheng F, Wu J, Liu B, Wang X. 2015. Genome evolution after whole genome triplication: the
subgenome dominance in Brassica rapa. In The Brassica rapa Genome, ed. X Wang, D Kole, pp. 107–14.
Berlin/Heidelberg: Springer
22. Chia J-M, Song C, Bradbury PJ, Costich D, de Leon N, et al. 2012. Maize HapMap2 identifies extant
variation from a genome in flux. Nat. Genet. 44(7):803–7
23. Clark RM, Tavaré S, Doebley J. 2005. Estimating a nucleotide substitution rate for maize from poly-
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

morphism at a major domestication locus. Mol. Biol. Evol. 22(11):2304–12

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

24. Clark RM, Wagler TN, Quijada P, Doebley J. 2006. A distant upstream enhancer at the maize domes-
tication gene tb1 has pleiotropic effects on plant and inflorescent architecture. Nat. Genet. 38(5):594–97
25. Cook BI, Smerdon JE, Seager R, Coats S. 2014. Global warming and 21st century drying. Clim. Dyn.
43(9–10):2607–27
26. Correns CE. 1900. G. Mendel’s Regel über das Verhalten der Nachkommenschaft der Rassenbastarde.
Ber. Dtsch. Bot. Ges. 18:158–67
27. Coumou D, Robinson A. 2013. Historic and future increase in the global land area affected by monthly
heat extremes. Environ. Res. Lett. 8(3):034018
28. Crossa J, Pérez-Rodrı́guez P, Cuevas J, Montesinos-López O, Jarquı́n D, et al. 2017. Genomic selection
in plant breeding: methods, models, and perspectives. Trends Plant Sci. 22(11):961–75
29. Curtin SJ, Tiffin P, Guhlin J, Trujillo DI, Burghardt LT, et al. 2017. Validating genome-wide association
candidates controlling quantitative variation in nodulation. Plant Physiol. 173(2):921–31
30. Daniel R. 2005. The metagenomics of soil. Nat. Rev. Microbiol. 3(6):470–78
31. Darwin C. 1868. The Variation of Animals and Plants Under Domestication. London: Murray
32. Darwin C. 1876. The Effects of Cross and Self-Fertilisation in the Vegetable Kingdom. London: Murray
33. de Quadros PD, Zhalnina K, Davis-Richardson A, Fagen JR, Drew J, et al. 2012. The effect of tillage
system and crop rotation on soil microbial diversity and composition in a subtropical Acrisol. Diversity
4(4):375–95
34. De Vries H. 1900. Sur la loi de disjonction des hybrides. Comptes Rendus Acad. Sci. 130:845–47
35. Denison RF. 2015. Evolutionary tradeoffs as opportunities to improve yield potential. Field Crops. Res.
182:3–8
36. Doebley JF, Gaut BS, Smith BD. 2006. The molecular genetics of crop domestication. Cell 127(7):1309–
21
37. Duvick DN. 2005. The contribution of breeding to yield advances in maize (Zea mays L.). Adv. Agron.
86:83–145
38. Dvorak J, Yang Z-L, You FM, Luo M-C. 2004. Deletion polymorphism in wheat chromosome regions
with contrasting recombination rates. Genetics 168(3):1665–75
39. Edger PP, Smith R, McKain MR, Cooley AM, Vallejo-Marin M, et al. 2017. Subgenome dominance in an
interspecific hybrid, synthetic allopolyploid, and a 140-year-old naturally established neo-allopolyploid
monkeyflower. Plant Cell 29(9):2150–67
40. Edwards J, Johnson C, Santos-Medellı́n C, Lurie E, Podishetty NK, et al. 2015. Structure, variation,
and assembly of the root-associated microbiomes of rice. PNAS 112(8):E911–20
41. Fasoula VA. 2013. Prognostic breeding: a new paradigm for crop improvement. In Plant Breeding Reviews,
ed. J Janick, pp. 297–347. Hoboken, NJ: Wiley
42. Felsenstein J. 1974. The evolutionary advantage of recombination. Genetics 78(2):737–56
43. Fisher RA. 1919. XV.—The correlation between relatives on the supposition of Mendelian inheritance.
Earth Environ. Sci. Trans. R. Soc. Edinb. 52(2):399–433

www.annualreviews.org • Crop Quantitative Genomics 437

 GE52CH19_Wallace ARI 17 October 2018 10:54

44. Fleischmann A, Michael TP, Rivadavia F, Sousa A, Wang W, et al. 2014. Evolution of genome size and
chromosome number in the carnivorous plant genus Genlisea (Lentibulariaceae), with a new estimate of
the minimum genome size in angiosperms. Ann. Bot. 114(8):1651–63
45. Fort A, Tuteja R, Braud M, McKeown PC, Spillane C. 2017. Parental-genome dosage effects on the
transcriptome of F1 hybrid triploid embryos of Arabidopsis thaliana. Plant J. 92(6):1044–58
46. Fu H, Dooner HK. 2002. Intraspecific violation of genetic colinearity and its implications in maize.
PNAS 99(14):9573–78
47. Galton F. 1886. Regression towards mediocrity in hereditary stature. J. Anthropol. Inst. Great Br. Ireland
15:246–63
48. Gardner CO, Lonnquist JH. 1959. Linkage and the degree of dominance of genes controlling quanti-
tative characters in maize. Agron. J. 51(9):524–28
49. Gdanetz K, Trail F. 2017. The wheat microbiome under four management strategies, and potential for
endophytes in disease protection. Phytobiomes 1(3):158–68
50. Gerland P, Raftery AE, Ševciková H, Li N, Gu D, et al. 2014. World population stabilization unlikely
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

this century. Science 346(6206):234–37

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

51. Gianola D. 2013. Priors in whole-genome regression: The Bayesian alphabet returns. Genetics
194(3):573–96
52. Glémin S. 2010. Surprising fitness consequences of GC-biased gene conversion: I. Mutation load and
inbreeding depression. Genetics 185(3):939–59
53. Graham GI, Wolff DW, Stuber CW. 1997. Characterization of a yield quantitative trait locus on chro-
mosome five of maize by fine mapping. Crop. Sci. 37:1601–10
54. Hanasaki N, Fujimori S, Yamamoto T, Yoshikawa S, Masaki Y, et al. 2013. A global water scarcity
assessment under shared socio-economic pathways—Part 2: Water availability and scarcity. Hydrol. Earth
Syst. Sci. 17:2393–413
55. Hancock AM, Brachi B, Faure N, Horton MW, Jarymowycz LB, et al. 2011. Adaptation to climate across
the Arabidopsis thaliana genome. Science 334(6052):83–86
56. Harjes CE, Rocheford TR, Bai L, Brutnell TP, Kandianis CB, et al. 2008. Natural genetic variation in
lycopene epsilon cyclase tapped for maize biofortification. Science 319(5861):330–33
57. Hartman K, van der Heijden MGA, Wittwer RA, Banerjee S, Walser J-C, Schlaeppi K. 2018. Cropping
practices manipulate abundance patterns of root and soil microbiome members paving the way to smart
farming. Microbiome 6(1):14
58. Hartmann M, Frey B, Mayer J, Mäder P, Widmer F. 2015. Distinct soil microbial diversity under
long-term organic and conventional farming. ISME J. 9(5):1177–94
59. He Z, Zhai W, Wen H, Tang T, Wang Y, et al. 2011. Two evolutionary histories in the genome of rice:
the roles of domestication genes. PLOS Genet. 7(6):e1002100
60. Hedden P. 2003. The genes of the Green Revolution. Trends Genet. 19(1):5–9
61. Heffner EL, Lorenz AJ, Jannink J-L, Sorrells ME. 2010. Plant breeding with genomic selection: gain
per unit time and cost. Crop. Sci. 50(5):1681–90
62. Heslot N, Yang H-P, Sorrells ME, Jannink J-L. 2012. Genomic selection in plant breeding: a comparison
of models. Crop. Sci. 52(1):146–60
63. Hickey JM, Chiurugwi T, Mackay I, Powell W, et al. (Implementing Genomic Selection in CGIAR
Breed. Programs Workshop Particip.). 2017. Genomic prediction unifies animal and plant breeding
programs to form platforms for biological discovery. Nat. Genet. 49(9):1297–303
64. Hill WG. 2014. Applications of population genetics to animal breeding, from Wright, Fisher and Lush
to genomic prediction. Genetics 196(1):1–16
65. Hill WG, Robertson A. 1966. The effect of linkage on limits to artificial selection. Genet. Res. 8(3):269–94
66. Hirakawa H, Okada Y, Tabuchi H, Shirasawa K, Watanabe A, et al. 2015. Survey of genome sequences
in a wild sweet potato, Ipomoea trifida (H.B.K.) G. Don. DNA Res. 22(2):171–79
67. Hirakawa H, Shirasawa K, Kosugi S, Tashiro K, Nakayama S, et al. 2014. Dissection of the octoploid
strawberry genome by deep sequencing of the genomes of Fragaria species. DNA Res. 21(2):169–81
68. Hirsch CN, Hirsch CD, Brohammer AB, Bowman MJ, Soifer I, et al. 2016. Draft assembly of elite inbred
line PH207 provides insights into genomic and transcriptome diversity in maize. Plant Cell 28(11):2700–
14

438 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

69. Huang L, Brooks SA, Li W, Fellers JP, Trick HN, Gill BS. 2003. Map-based cloning of leaf rust resistance
gene Lr21 from the large and polyploid genome of bread wheat. Genetics 164(2):655–64
70. Huang X, Yang S, Gong J, Zhao Y, Feng Q, et al. 2015. Genomic analysis of hybrid rice varieties reveals
numerous superior alleles that contribute to heterosis. Nat. Commun. 6:6258
71. Huang Y-F, Poland JA, Wight CP, Jackson EW, Tinker NA. 2014. Using genotyping-by-sequencing
(GBS) for genomic discovery in cultivated oat. PLOS ONE 9(7):e102448
72. Hufford MB, Xu X, van Heerwaarden J, Pyhäjärvi T, Chia J-M, et al. 2012. Comparative population
genomics of maize domestication and improvement. Nat. Genet. 44(7):808–11
73. Hung H-Y, Shannon LM, Tian F, Bradbury PJ, Chen C, et al. 2012. ZmCCT and the genetic basis of
day-length adaptation underlying the postdomestication spread of maize. PNAS 109(28):E1913–21
74. Innan H, Kim Y. 2004. Pattern of polymorphism after strong artificial selection in a domestication event.
PNAS 101(29):10667–72
75. Int. Rice Genome Seq. Proj. 2005. The map-based sequence of the rice genome. Nature 436(7052):793–
800
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

76. Jacobson A, Lian L, Zhong S, Bernardo R. 2014. General combining ability model for genomewide
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

selection in a biparental cross. Crop. Sci. 54:895–905

77. Jiang Y, Li S, Li R, Zhang J, Liu Y, et al. 2017. Plant cultivars imprint the rhizosphere bacterial community
composition and association networks. Soil Biol. Biochem. 109:145–55
78. Jiang Y, Schmidt RH, Zhao Y, Reif JC. 2017. A quantitative genetic framework highlights the role of
epistatic effects for grain-yield heterosis in bread wheat. Nat. Genet. 49(12):1741–46
79. Jiao Y, Wickett NJ, Ayyampalayam S, Chanderbali AS, Landherr L, et al. 2011. Ancestral polyploidy in
seed plants and angiosperms. Nature 473(7345):97–100
80. Kamfwa K, Cichy KA, Kelly JD. 2015. Genome-wide association analysis of symbiotic nitrogen fixation
in common bean. Theor. Appl. Genet. 128(10):1999–2017
81. Khoshbakht K, Hammer K. 2008. How many plant species are cultivated? Genet. Resour. Crop Evol.
55(7):925–28
82. Khoury CK, Achicanoy HA, Bjorkman AD, Navarro-Racines C, Guarino L, et al. 2016. Origins of food
crops connect countries worldwide. Proc. R. Soc. B. 283(1832):20160792
83. Kirkpatrick M, Jarne P. 2000. The effects of a bottleneck on inbreeding depression and the genetic load.
Am. Nat. 155(2):154–67
84. Koenig D, Jiménez-Gómez JM, Kimura S, Fulop D, Chitwood DH, et al. 2013. Comparative transcrip-
tomics reveals patterns of selection in domesticated and wild tomato. PNAS 110(28):E2655–62
85. Komatsuda T, Pourkheirandish M, He C, Azhaguvel P, Kanamori H, et al. 2007. Six-rowed barley orig-
inated from a mutation in a homeodomain-leucine zipper I-class homeobox gene. PNAS 104(4):1424–29
86. Kovach MJ, Calingacion MN, Fitzgerald MA, McCouch SR. 2009. The origin and evolution of fragrance
in rice (Oryza sativa L.). PNAS 106(34):14444–49
87. Kover PX, Valdar W, Trakalo J, Scarcelli N, Ehrenreich IM, et al. 2009. A multiparent advanced
generation inter-cross to fine-map quantitative traits in Arabidopsis thaliana. PLOS Genet. 5(7):e1000551
88. Krasileva KV, Vasquez-Gross HA, Howell T, Bailey P, Paraiso F, et al. 2017. Uncovering hidden
variation in polyploid wheat. PNAS 114(6):E913–21
89. Krchov L-M, Bernardo R. 2015. Relative efficiency of genomewide selection for testcross performance
of doubled haploid lines in a maize breeding program. Crop. Sci. 55:2091–99
90. Lake L, Li Y, Casal JJ, Sadras VO. 2016. Negative association between chickpea response to competition
and crop yield: phenotypic and genetic analysis. Field Crops. Res. 196:409–17
91. Lander ES, Botstein D. 1989. Mapping Mendelian factors underlying quantitative traits using RFLP
linkage maps. Genetics 121(1):185–99
92. Larièpe A, Mangin B, Jasson S, Combes V, Dumas F, et al. 2011. The genetic basis of heterosis:
multiparental quantitative trait loci mapping reveals contrasted levels of apparent overdominance among
traits of agronomical interest in maize (Zea mays L.). Genetics 190(2):795–811
93. Lasky JR, Des Marais DL, Lowry DB, Povolotskaya I, McKay JK, et al. 2014. Natural variation in abiotic
stress responsive gene expression and local adaptation to climate in Arabidopsis thaliana. Mol. Biol. Evol.
31(9):2283–96

www.annualreviews.org • Crop Quantitative Genomics 439

 GE52CH19_Wallace ARI 17 October 2018 10:54

94. Lasky JR, Upadhyaya HD, Ramu P, Deshpande S, Hash CT, et al. 2015. Genome-environment associ-
ations in sorghum landraces predict adaptive traits. Sci. Adv. 1(6):e1400218
95. Leiser WL, Olatoye MO, Rattunde HFW, Neumann G, Weltzien E, Haussmann BIG. 2016. No need
to breed for enhanced colonization by arbuscular mycorrhizal fungi to improve low-P adaptation of West
African sorghums. Plant Soil 401(1–2):51–64
96. Li C, Sun B, Li Y, Liu C, Wu X, et al. 2016. Numerous genetic loci identified for drought tolerance in
the maize nested association mapping populations. BMC Genomics 17(1):894
97. Lian L, Jacobson A, Zhong S, Bernardo R. 2014. Genomewide prediction accuracy within 969 maize
biparental populations. Crop. Sci. 54:1514–22
98. Li X, Li X, Fridman E, Tesso TT, Yu J. 2015. Dissecting repulsion linkage in the dwarfing gene Dw3
region for sorghum plant height provides insights into heterosis. PNAS 112(38):11823–28
99. Li Y, Xiao J, Wu J, Duan J, Liu Y, et al. 2012. A tandem segmental duplication (TSD) in green revolution
gene Rht-D1b region underlies plant height variation. New Phytol. 196(1):282–91
100. Li Y-H, Zhao S-C, Ma J-X, Li D, Yan L, et al. 2013. Molecular footprints of domestication and im-
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

provement in soybean revealed by whole genome re-sequencing. BMC Genomics 14:579

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

101. Li Z, Coffey L, Garfin J, Miller ND, White MR, et al. 2018. Genotype-by-environment interactions
affecting heterosis in maize. PLOS ONE 13(1):e0191321
102. Lin T, Zhu G, Zhang J, Xu X, Yu Q, et al. 2014. Genomic analyses provide insights into the history of
tomato breeding. Nat. Genet. 46(11):1220–26
103. Lin Z, Li X, Shannon LM, Yeh C-T, Wang ML, et al. 2012. Parallel domestication of the Shattering1
genes in cereals. Nat. Genet. 44(6):720–24
104. Lu F, Romay MC, Glaubitz JC, Bradbury PJ, Elshire RJ, et al. 2015. High-resolution genetic mapping
of maize pan-genome sequence anchors. Nat. Commun. 6:6914
105. Lu J, Tang T, Tang H, Huang J, Shi S, Wu C-I. 2006. The accumulation of deleterious mutations in
rice genomes: a hypothesis on the cost of domestication. Trends Genet. 22(3):126–31
106. Lundberg DS, Lebeis SL, Paredes SH, Yourstone S, Gehring J, et al. 2012. Defining the core Arabidopsis
thaliana root microbiome. Nature 488(7409):86–90
107. Mace ES, Hunt CH, Jordan DR. 2013. Supermodels: sorghum and maize provide mutual insight into
the genetics of flowering time. Theor. Appl. Genet. 126(5):1377–95
108. Madlung A. 2013. Polyploidy and its effect on evolutionary success: old questions revisited with new
tools. Heredity 110(2):99–104
109. Maron LG, Guimarães CT, Kirst M, Albert PS, Birchler JA, et al. 2013. Aluminum tolerance in maize
is associated with higher MATE1 gene copy number. PNAS 110(13):5241–46
110. Marsden CD, Ortega-Del Vecchyo D, O’Brien DP, Taylor JF, Ramirez O, et al. 2016. Bottlenecks
and selective sweeps during domestication have increased deleterious genetic variation in dogs. PNAS
113(1):152–57
111. Martin GB, Brommonschenkel SH, Chunwongse J, Frary A, Ganal MW, et al. 1993. Map-based cloning
of a protein kinase gene conferring disease resistance in tomato. Science 262(5138):1432–36
112. Martinez Barrio A, Lamichhaney S, Fan G, Rafati N, Pettersson M, et al. 2016. The genetic basis for
ecological adaptation of the Atlantic herring revealed by genome sequencing. eLife 5:e12081
113. Mathew RP, Feng Y, Githinji L, Ankumah R, Balkcom KS. 2012. Impact of no-tillage and conventional
tillage systems on soil microbial communities. Appl. Environ. Soil Sci. 2012:548620
114. Maurer A, Draba V, Jiang Y, Schnaithmann F, Sharma R, et al. 2015. Modelling the genetic architecture
of flowering time control in barley through nested association mapping. BMC Genomics 16:290
115. Mayrose I, Zhan SH, Rothfels CJ, Magnuson-Ford K, Barker MS, et al. 2011. Recently formed polyploid
plants diversify at lower rates. Science 333(6047):1257
116. McCouch SR, Wright MH, Tung C-W, Maron LG, McNally KL, et al. 2016. Open access resources
for genome-wide association mapping in rice. Nat. Commun. 7:10532
117. McMullen MD, Kresovich S, Villeda HS, Bradbury P, Li H, et al. 2009. Genetic properties of the maize
nested association mapping population. Science 325(5941):737–40
118. Mendes R, Kruijt M, de Bruijn I, Dekkers E, van der Voort M, et al. 2011. Deciphering the rhizosphere
microbiome for disease-suppressive bacteria. Science 332(6033):1097–100

440 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

119. Meuwissen TH, Hayes BJ, Goddard ME. 2001. Prediction of total genetic value using genome-wide
dense marker maps. Genetics 157(4):1819–29
120. Meyer RS, DuVal AE, Jensen HR. 2012. Patterns and processes in crop domestication: an historical
review and quantitative analysis of 203 global food crops. New Phytol. 196(1):29–48
121. Moll RH, Lindsey MF, Robinson HF. 1964. Estimates of genetic variances and level of dominance in
maize. Genetics 49(3):411–23
122. Morris GP, Ramu P, Deshpande SP, Hash CT, Shah T, et al. 2013. Population genomic and genome-
wide association studies of agroclimatic traits in sorghum. PNAS 110(2):453–58
123. Motamayor JC, Mockaitis K, Schmutz J, Haiminen N, Livingstone D 3rd, et al. 2013. The genome
sequence of the most widely cultivated cacao type and its use to identify candidate genes regulating pod
color. Genome Biol. 14(6):r53
124. Müller NA, Wijnen CL, Srinivasan A, Ryngajllo M, Ofner I, et al. 2016. Domestication selected for
deceleration of the circadian clock in cultivated tomato. Nat. Genet. 48(1):89–93
125. Multani DS, Briggs SP, Chamberlin MA, Blakeslee JJ, Murphy AS, Johal GS. 2003. Loss of an MDR
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

transporter in compact stalks of maize br2 and sorghum dw3 mutants. Science 302(5642):81–84
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

126. Nabholz B, Sarah G, Sabot F, Ruiz M, Adam H, et al. 2014. Transcriptome population genomics reveals
severe bottleneck and domestication cost in the African rice (Oryza glaberrima). Mol. Ecol. 23(9):2210–27
127. Nasseer AM, Martin JM, Heo HY, Blake NK, Sherman JD, et al. 2016. Impact of a quantitative trait
locus for tiller number on plasticity of agronomic traits in spring wheat. Crop. Sci. 56:595–602
128. Orellana LH, Chee-Sanford JC, Sanford RA, Löffler FE, Konstantinidis KT. 2017. Year-round shotgun
metagenomes reveal stable microbial communities in agricultural soils and novel ammonia oxidizers
responding to fertilization. Appl. Environ. Microbiol. 84:e01646-17
129. Orr HA. 1998. The population genetics of adaptation: the distribution of factors fixed during adaptive
evolution. Evolution 52(4):935–49
130. Orr HA. 2000. Adaptation and the cost of complexity. Evolution 54(1):13–20
131. Ossowski S, Schneeberger K, Lucas-Lledó JI, Warthmann N, Clark RM, et al. 2010. The rate and
molecular spectrum of spontaneous mutations in Arabidopsis thaliana. Science 327(5961):92–94
132. Panke-Buisse K, Poole AC, Goodrich JK, Ley RE, Kao-Kniffin J. 2015. Selection on soil microbiomes
reveals reproducible impacts on plant function. ISME J. 9:980–89
133. Pascual L, Desplat N, Huang BE, Desgroux A, Bruguier L, et al. 2015. Potential of a tomato MAGIC
population to decipher the genetic control of quantitative traits and detect causal variants in the rese-
quencing era. Plant Biotechnol. J. 13(4):565–77
134. Peiffer JA, Romay MC, Gore MA, Flint-Garcia SA, Zhang Z, et al. 2014. The genetic architecture of
maize height. Genetics 196(4):1337–56
135. Peiffer JA, Spor A, Koren O, Jin Z, Tringe SG, et al. 2013. Diversity and heritability of the maize
rhizosphere microbiome under field conditions. PNAS 110(16):6548–53
136. Pellicer J, Fay MF, Leitch IJ. 2010. The largest eukaryotic genome of them all? Bot. J. Linn. Soc.
164(1):10–15
137. Price AL, Zaitlen NA, Reich D, Patterson N. 2010. New approaches to population stratification in
genome-wide association studies. Nat. Rev. Genet. 11(7):459–63
138. Prochnik S, Marri PR, Desany B, Rabinowicz PD, Kodira C, et al. 2012. The cassava genome: current
progress, future directions. Trop. Plant Biol. 5(1):88–94
139. Purugganan MD, Boyles AL, Suddith JI. 2000. Variation and selection at the CAULIFLOWER floral
homeotic gene accompanying the evolution of domesticated Brassica oleracea. Genetics 155(2):855–62
140. Ralph PL, Coop G. 2015. Convergent evolution during local adaptation to patchy landscapes. PLOS
Genet. 11(11):e1005630
141. Ramu P, Esuma W, Kawuki R, Rabbi IY, Egesi C, et al. 2016. Cassava HapMap: Masking deleterious
mutations in a clonal crop species. BioRxiv 077123. https://doi.org/10.1101/077123
142. Ray DK, Mueller ND, West PC, Foley JA. 2013. Yield trends are insufficient to double global crop
production by 2050. PLOS ONE 8(6):e66428
143. Reams AB, Roth JR. 2015. Mechanisms of gene duplication and amplification. Cold Spring Harb. Perspect.
Biol. 7(2):a016592

www.annualreviews.org • Crop Quantitative Genomics 441

 GE52CH19_Wallace ARI 17 October 2018 10:54

144. Renaut S, Rieseberg LH. 2015. The accumulation of deleterious mutations as a consequence of domes-
tication and improvement in sunflowers and other Compositae crops. Mol. Biol. Evol. 32(9):2273–83
145. Renny-Byfield S, Rodgers-Melnick E, Ross-Ibarra J. 2016. Gene fractionation and function in the ancient
subgenomes of maize. Mol. Biol. Evol. 34(8):1825–32
146. Riedelsheimer C, Melchinger AE. 2013. Optimizing the allocation of resources for genomic selection
in one breeding cycle. Theor. Appl. Genet. 126(11):2835–48
147. Risch N, Merikangas K. 1996. The future of genetic studies of complex human diseases. Science
273(5281):1516–17
148. Rodgers-Melnick E, Bradbury PJ, Elshire RJ, Glaubitz JC, Acharya CB, et al. 2015. Recombination in
diverse maize is stable, predictable, and associated with genetic load. PNAS 112(12):3823–28
149. Rodgers-Melnick E, Vera DL, Bass HW, Buckler ES. 2016. Open chromatin reveals the functional
maize genome. PNAS 113(22):E3177–84
150. Saade S, Maurer A, Shahid M, Oakey H, Schmöckel SM, et al. 2016. Yield-related salinity tolerance
traits identified in a nested association mapping (NAM) population of wild barley. Sci. Rep. 6:32586
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

151. Sadras VO, Lawson C, Montoro A. 2012. Photosynthetic traits in Australian wheat varieties released
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

between 1958 and 2007. Field Crops. Res. 134:19–29

152. Sallam A, Martsch R. 2015. Association mapping for frost tolerance using multi-parent advanced gen-
eration inter-cross (MAGIC) population in faba bean (Vicia faba L.). Genetica 143(4):501–14
153. Salvi S, Sponza G, Morgante M, Tomes D, Niu X, et al. 2007. Conserved noncoding genomic sequences
associated with a flowering-time quantitative trait locus in maize. PNAS 104(27):11376–81
154. SanMiguel P, Gaut BS, Tikhonov A, Nakajima Y, Bennetzen JL. 1998. The paleontology of intergene
retrotransposons of maize. Nat. Genet. 20(1):43–45
155. Sannemann W, Huang BE, Mathew B, Léon J. 2015. Multi-parent advanced generation inter-cross in
barley: high-resolution quantitative trait locus mapping for flowering time as a proof of concept. Mol.
Breed. 35:86
156. Schnable JC, Springer NM, Freeling M. 2011. Differentiation of the maize subgenomes by genome
dominance and both ancient and ongoing gene loss. PNAS 108(10):4069–74
157. Schnable PS, Springer NM. 2013. Progress toward understanding heterosis in crop plants. Annu. Rev.
Plant Biol. 64:71–88
158. Schön CC, Dhillon BS, Utz HF, Melchinger AE. 2009. High congruency of QTL positions for heterosis
of grain yield in three crosses of maize. Theor. Appl. Genet. 120(2):321–32
159. Schubert M, Jónsson H, Chang D, Der Sarkissian C, Ermini L, et al. 2014. Prehistoric genomes reveal
the genetic foundation and cost of horse domestication. PNAS 111(52):E5661–69
160. Sexton JP, Hangartner SB, Hoffmann AA. 2014. Genetic isolation by environment or distance: Which
pattern of gene flow is most common? Evolution 68(1):1–15
161. Shulaev V, Sargent DJ, Crowhurst RN, Mockler TC, Folkerts O, et al. 2011. The genome of woodland
strawberry (Fragaria vesca). Nat. Genet. 43(2):109–16
162. Soman C, Li D, Wander MM, Kent AD. 2016. Long-term fertilizer and crop-rotation treatments
differentially affect soil bacterial community structure. Plant Soil 413(1–2):145–59
163. Song WY, Wang GL, Chen LL, Kim HS, Pi LY, et al. 1995. A receptor kinase-like protein encoded by
the rice disease resistance gene, Xa21. Science 270(5243):1804–6
164. Stanton-Geddes J, Paape T, Epstein B, Briskine R, Yoder J, et al. 2013. Candidate genes and genetic
architecture of symbiotic and agronomic traits revealed by whole-genome, sequence-based association
genetics in Medicago truncatula. PLOS ONE 8(5):e65688
165. Studer AJ, Doebley JF. 2011. Do large effect QTL fractionate? A case study at the maize domestication
QTL teosinte branched1. Genetics 188(3):673–81
166. Sukumaran S, Reynolds MP, Lopes MS, Crossa J. 2015. Genome-wide association study for adaptation
to agronomic plant density: a component of high yield potential in spring wheat. Crop. Sci. 55:2609–19
167. Taketa S, Amano S, Tsujino Y, Sato T, Saisho D, et al. 2008. Barley grain with adhering hulls is
controlled by an ERF family transcription factor gene regulating a lipid biosynthesis pathway. PNAS
105(10):4062–67
168. Tan B-C, Guan J-C, Ding S, Wu S, Saunders JW, et al. 2017. Structure and origin of the White Cap
locus and its role in evolution of grain color in maize. Genetics 206(1):135–50

442 Wallace · Rodgers-Melnick · Buckler

 GE52CH19_Wallace ARI 17 October 2018 10:54

169. Technow F, Schrag TA, Schipprack W, Bauer E, Simianer H, Melchinger AE. 2014. Genome properties
and prospects of genomic prediction of hybrid performance in a breeding program of maize. Genetics
197(4):1343–55
170. Thompson LR, Sanders JG, McDonald D, Amir A, Ladau J, et al. 2017. A communal catalogue reveals
Earth’s multiscale microbial diversity. Nature 551:457–63
171. Tilman D, Clark M. 2014. Global diets link environmental sustainability and human health. Nature
515(7528):518–22
172. Tracy WF, Whitt SR, Buckler ES. 2006. Recurrent mutation and genome evolution: example of and the
origin of sweet maize. Crop. Sci. 46(Suppl. 1):S49–54
173. Tschermak E. 1900. Ueber künstsliche Kreuzung bei Pisum sativum. In Berichte der Deutsche Botanischen
Gessellschaft, pp. 232–39. Berlin: Gebrüder Borntræger
174. Turner TR, Ramakrishnan K, Walshaw J, Heavens D, Alston M, et al. 2013. Comparative metatranscrip-
tomics reveals kingdom level changes in the rhizosphere microbiome of plants. ISME J. 7(12):2248–58
175. Upadhyaya HD, Vetriventhan M, Deshpande SP, Sivasubramani S, Wallace JG, et al. 2015. Population
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

genetics and structure of a global foxtail millet germplasm collection. Plant Genome 8:3
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

176. Varshney RK, Roorkiwal M, Sorrells ME, eds. 2017. Genomic Selection for Crop Improvement: New Molec-
ular Breeding Strategies for Crop Improvement. Cham, Switz.: Springer
177. Varshney RK, Shi C, Thudi M, Mariac C, Wallace J, et al. 2017. Pearl millet genome sequence provides
a resource to improve agronomic traits in arid environments. Nat. Biotechnol. 35(10):969–76
178. Venuprasad R, Bool ME, Quiatchon L, Sta Cruz MT, Amante M, Atlin GN. 2012. A large-effect QTL
for rice grain yield under upland drought stress on chromosome 1. Mol. Breed. 30(1):535–47
179. Walker V, Bertrand C, Bellvert F, Moënne-Loccoz Y, Bally R, Comte G. 2011. Host plant secondary
metabolite profiling shows a complex, strain-dependent response of maize to plant growth-promoting
rhizobacteria of the genus Azospirillum. New Phytol. 189(2):494–506
180. Wallace JG, Bradbury PJ, Zhang N, Gibon Y, Stitt M, Buckler ES. 2014. Association mapping across
numerous traits reveals patterns of functional variation in maize. PLOS Genet. 10(12):e1004845
181. Wallace JG, Upadhyaya HD, Vetriventhan M, Buckler ES, Hash CT, Ramu P. 2015. The genetic
makeup of a global barnyard millet germplasm collection. Plant Genome 8:1
182. Walley JW, Sartor RC, Shen Z, Schmitz RJ, Wu KJ, et al. 2016. Integration of omic networks in a
developmental atlas of maize. Science 353(6301):814–18
183. Wang H, Nussbaum-Wagler T, Li B, Zhao Q, Vigouroux Y, et al. 2005. The origin of the naked grains
of maize. Nature 436(7051):714–19
184. Wang Z, Liu L, Chen Q, Wen X, Liao Y. 2016. Conservation tillage increases soil bacterial diversity in
the dryland of northern China. Agron. Sustain. Dev. 36(2):28
185. Weber B, Zicola J, Oka R, Stam M. 2016. Plant enhancers: a call for discovery. Trends Plant Sci.
21(11):974–87
186. Wendel JF. 2015. The wondrous cycles of polyploidy in plants. Am. J. Bot. 102(11):1753–56
187. World Bank. 2016. World development indicators 2016. World Bank Group. http://documents.
worldbank.org/curated/en/805371467990952829/World-development-indicators-2016
188. Wright S. 1921. Systems of mating. I. The biometric relations between parent and offspring. Genetics
6(2):111–23
189. Xu K, Xu X, Fukao T, Canlas P, Maghirang-Rodriguez R, et al. 2006. Sub1A is an ethylene-response-
factor-like gene that confers submergence tolerance to rice. Nature 442(7103):705–8
190. Xu X, Liu X, Ge S, Jensen JD, Hu F, et al. 2011. Resequencing 50 accessions of cultivated and wild rice
yields markers for identifying agronomically important genes. Nat. Biotechnol. 30(1):105–11
191. Yahiaoui N, Srichumpa P, Dudler R, Keller B. 2004. Genome analysis at different ploidy levels allows
cloning of the powdery mildew resistance gene Pm3b from hexaploid wheat. Plant J. 37(4):528–38
192. Yan J, Kandianis CB, Harjes CE, Bai L, Kim E-H, et al. 2010. Rare genetic variation at Zea mays crtRB1
increases β-carotene in maize grain. Nat. Genet. 42(4):322–27
193. Yang J, Mezmouk S, Baumgarten A, Buckler ES, Guill KE, et al. 2017. Incomplete dominance of deleteri-
ous alleles contributes substantially to trait variation and heterosis in maize. PLOS Genet. 13(9):e1007019
194. Yao H, Dogra Gray A, Auger DL, Birchler JA. 2013. Genomic dosage effects on heterosis in triploid
maize. PNAS 110(7):2665–69

www.annualreviews.org • Crop Quantitative Genomics 443

 GE52CH19_Wallace ARI 17 October 2018 10:54

195. Yeaman S, Hodgins KA, Lotterhos KE, Suren H, Nadeau S, et al. 2016. Convergent local adaptation to
climate in distantly related conifers. Science 353(6306):1431–33
196. York LM, Galindo-Castañeda T, Schussler JR, Lynch JP. 2015. Evolution of US maize (Zea mays L.)
root architectural and anatomical phenes over the past 100 years corresponds to increased tolerance of
nitrogen stress. J. Exp. Bot. 66(8):2347–58
197. York LM, Lynch JP. 2015. Intensive field phenotyping of maize (Zea mays L.) root crowns identi-
fies phenes and phene integration associated with plant growth and nitrogen acquisition. J. Exp. Bot.
66(18):5493–505
198. Zabel F, Putzenlechner B, Mauser W. 2014. Global agricultural land resources–a high resolution suitabil-
ity evaluation and its perspectives until 2100 under climate change conditions. PLOS ONE 9(9):e107522
199. Zhang W, Wu Y, Schnable JC, Zeng Z, Freeling M, et al. 2012. High-resolution mapping of open
chromatin in the rice genome. Genome Res. 22(1):151–62
200. Zhang Y, Liu Z, Khan AA, Lin Q, Han Y, et al. 2016. Expression partitioning of homeologs and tandem
duplications contribute to salt tolerance in wheat (Triticum aestivum L.). Sci. Rep. 6:21476
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

201. Zhao Y, Li Z, Liu G, Jiang Y, Maurer HP, et al. 2015. Genome-based establishment of a high-yielding
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

heterotic pattern for hybrid wheat breeding. PNAS 112(51):15624–29

202. Zhou Z, Jiang Y, Wang Z, Gou Z, Lyu J, et al. 2015. Resequencing 302 wild and cultivated accessions
identifies genes related to domestication and improvement in soybean. Nat. Biotechnol. 33(4):408–14
203. Zhu X-G, Lynch JP, LeBauer DS, Millar AJ, Stitt M, Long SP. 2016. Plants in silico: Why, why now
and what?–an integrative platform for plant systems biology research. Plant Cell Environ. 39(5):1049–57

444 Wallace · Rodgers-Melnick · Buckler

 GE52-FrontMatter ARI 17 October 2018 11:38

Annual Review of
Genetics

Volume 52, 2018

Contents

Tracing My Roots: How I Became a Plant Biologist

Frederick M. Ausubel p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Transgenerational Epigenetic Inheritance
Ana Bošković and Oliver J. Rando p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p21
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.
Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Mechanisms of Neural Crest Migration

András Szabó and Roberto Mayor p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p43
The Hippo Signaling Network and Its Biological Functions
Jyoti R. Misra and Kenneth D. Irvine p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p65
The Smc5/6 Complex: New and Old Functions of the Enigmatic
Long-Distance Relative
Luis Aragón p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p89
H3–H4 Histone Chaperone Pathways
Prerna Grover, Jonathon S. Asa, and Eric I. Campos p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 109
piRNA-Guided Genome Defense: From Biogenesis to Silencing
Benjamin Czech, Marzia Munafò, Filippo Ciabrelli, Exvelyn L. Eastwood,
Martin H. Fabry, Emma Kneuss, and Gregory J. Hannon p p p p p p p p p p p p p p p p p p p p p p p p p p p p 131
Metabolic Gene Clusters in Eukaryotes
Hans-Wilhelm Nützmann, Claudio Scazzocchio, and Anne Osbourn p p p p p p p p p p p p p p p p p p p 159
Genetic Control of Early Cell Lineages in the Mammalian Embryo
Janet Rossant p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 185
Power in Numbers: Single-Cell RNA-Seq Strategies to Dissect
Complex Tissues
Kenneth D. Birnbaum p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 203
Shelterin-Mediated Telomere Protection
Titia de Lange p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 223
Understanding the Genetic Basis of C4 Kranz Anatomy with a View to
Engineering C3 Crops
Olga V. Sedelnikova, Thomas E. Hughes, and Jane A. Langdale p p p p p p p p p p p p p p p p p p p p p p p p 249
Aging in a Dish: iPSC-Derived and Directly Induced Neurons for
Studying Brain Aging and Age-Related Neurodegenerative Diseases
Jerome Mertens, Dylan Reid, Shong Lau, Yongsung Kim, and Fred H. Gage p p p p p p p p p p 271

viii
 GE52-FrontMatter ARI 17 October 2018 11:38

Chromosome Dynamics in Response to DNA Damage

Andrew Seeber, Michael H. Hauer, and Susan M. Gasser p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 295
Ribosome Hibernation
Thomas Prossliner, Kristoffer Skovbo Winther, Michael Askvad Sørensen,
and Kenn Gerdes p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 321
Chemical Modifications in the Life of an mRNA Transcript
Sigrid Nachtergaele and Chuan He p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 349
Calcium Channelopathies: Structural Insights into Disorders of the
Muscle Excitation–Contraction Complex
Raika Pancaroglu and Filip Van Petegem p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 373
Access provided by Universidade Federal de Lavras on 04/24/21. For personal use only.

Somatic Mutagenesis in Mammals and Its Implications for Human

Annu. Rev. Genet. 2018.52:421-444. Downloaded from www.annualreviews.org

Disease and Aging

Lei Zhang and Jan Vijg p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 397
On the Road to Breeding 4.0: Unraveling the Good, the Bad, and the
Boring of Crop Quantitative Genomics
Jason G. Wallace, Eli Rodgers-Melnick, and Edward S. Buckler p p p p p p p p p p p p p p p p p p p p p p p p 421
Phage-Encoded Anti-CRISPR Defenses
Sabrina Y. Stanley and Karen L. Maxwell p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 445
Unique Archaeal Small RNAs
José Vicente Gomes-Filho, Michael Daume, and Lennart Randau p p p p p p p p p p p p p p p p p p p p p p p 465
Recent Advances in Behavioral (Epi)Genetics in Eusocial Insects
Comzit Opachaloemphan, Hua Yan, Alexandra Leibholz, Claude Desplan,
and Danny Reinberg p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 489
The Multiple Levels of Mitonuclear Coregulation
R. Stefan Isaac, Erik McShane, and L. Stirling Churchman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 511
X-Chromosome Inactivation: A Crossroads Between Chromosome
Architecture and Gene Regulation
Rafael Galupa and Edith Heard p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 535
Immunoglobulin-Like Receptors and Their Impact on Wiring of Brain
Synapses
Scott Cameron and A. Kimberley McAllister p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 567

Errata

An online log of corrections to Annual Review of Genetics articles may be found at

http://www.annualreviews.org/errata/genet

Contents ix