Although applied over extremely short timescales, artificial selection has dramatically altered the form, physiology, and life history of cultivated plants. We have used RNAseq to define both gene sequence and expression divergence between cultivated tomato and five related wild species. Based on sequence differences, we detect footprints of positive selection in over 50 genes. We also document thousands of shifts in gene-expression level, many of which resulted from changes in selection pressure. These rapidly evolving genes are commonly associated with environmental response and stress tolerance. The importance of environmental inputs during evolution of gene expression is further highlighted by large-scale alteration of the light response coexpression network between wild and cultivated accessions. Human manipulation of the genome has heavily impacted the tomato transcriptome through directed admixture and by indirectly favoring nonsynonymous over synonymous substitutions. Taken together, our results shed light on the pervasive effects artificial and natural selection have had on the transcriptomes of tomato and its wild relatives.domestication | biotic stress | abiotic stress D omestication has long served as an important example of severe phenotypic divergence in response to selection. Darwin recognized the parallel between the processes of domestication and adaptation in the wild and used this analogy to emphasize the power of selection in generating phenotypic diversity (1). The genetic basis of domestication-associated phenotypes has been examined in several instances, most notably in maize, rice, tomato, and dogs (reviewed in refs. 2-5). The clear conclusion from these studies is that the rapid phenotypic divergence associated with domestication is often attributable to very few genetic loci (6). Improvements to DNA sequence technologies have allowed studies of the effect of domestication at the whole-genome level. Early population genetic analyses in maize found that very few genes (∼5%) show evidence of positive selection during domestication of maize (7), and recent work using whole-genome resequencing has found a similar proportion of the genome was under positive selection (8). Evidence for strong selective sweeps at a limited number of loci has also been found in rice and dog genomes (9). Together with the previous genetic mapping work, these studies support the model that relatively few mutations experienced extremely strong selection by humans during domestication.Although not the target of direct positive selection, the rest of the genome still experiences a dramatic shift in evolutionary pressures during domestication. Most characterized domestication events are associated with an extreme genetic bottleneck and alleviation of selective constraints in the original niche (10). These factors are predicted to increase the relative rate of nonsynonymous to synonymous (dN/dS) substitution, potentially resulting in the fixation of deleterious alleles (11). Previous studies comparing the distribution ...

Introgression lines (ILs), in which genetic material from wild tomato species is introgressed into a domesticated background, have been used extensively in tomato (Solanum lycopersicum) improvement. Here, we genotype an IL population derived from the wild desert tomato Solanum pennellii at ultrahigh density, providing the exact gene content harbored by each line. To take advantage of this information, we determine IL phenotypes for a suite of vegetative traits, ranging from leaf complexity, shape, and size to cellular traits, such as stomatal density and epidermal cell phenotypes. Elliptical Fourier descriptors on leaflet outlines provide a global analysis of highly heritable, intricate aspects of leaf morphology. We also demonstrate constraints between leaflet size and leaf complexity, pavement cell size, and stomatal density and show independent segregation of traits previously assumed to be genetically coregulated. Meta-analysis of previously measured traits in the ILs shows an unexpected relationship between leaf morphology and fruit sugar levels, which RNA-Seq data suggest may be attributable to genetically coregulated changes in fruit morphology or the impact of leaf shape on photosynthesis. Together, our results both improve upon the utility of an important genetic resource and attest to a complex, genetic basis for differences in leaf morphology between natural populations.

Modern systems biology permits the study of complex networks, such as circadian clocks, and the use of complex methodologies, such as quantitative genetics. However, it is difficult to combine these approaches due to factorial expansion in experiments when networks are examined using complex methods. We developed a genomic quantitative genetic approach to overcome this problem, allowing us to examine the function(s) of the plant circadian clock in different populations derived from natural accessions. Using existing microarray data, we defined 24 circadian time phase groups (i.e., groups of genes with peak phases of expression at particular times of day). These groups were used to examine natural variation in circadian clock function using existing single time point microarray experiments from a recombinant inbred line population. We identified naturally variable loci that altered circadian clock outputs and linked these circadian quantitative trait loci to preexisting metabolomics quantitative trait loci, thereby identifying possible links between clock function and metabolism. Using single-gene isogenic lines, we found that circadian clock output was altered by natural variation in Arabidopsis thaliana secondary metabolism. Specifically, genetic manipulation of a secondary metabolic enzyme led to altered free-running rhythms. This represents a unique and valuable approach to the study of complex networks using quantitative genetics.

Leaf shape is mutable, changing in ways modulated by both development and environment within genotypes. A complete model of leaf phenotype would incorporate the changes in leaf shape during juvenile-to-adult phase transitions and the ontogeny of each leaf. Here, we provide a morphometric description of >33,000 leaflets from a set of tomato (Solanum spp) introgression lines grown under controlled environment conditions. We first compare the shape of these leaves, arising during vegetative development, with >11,000 previously published leaflets from a field setting and >11,000 leaflets from wild tomato relatives. We then quantify the changes in shape, across ontogeny, for successive leaves in the heteroblastic series. Using principal component analysis, we then separate genetic effects modulating (1) the overall shape of all leaves versus (2) the shape of specific leaves in the series, finding the former more heritable than the latter and comparing quantitative trait loci regulating each. Our results demonstrate that phenotype is highly contextual and that unbiased assessments of phenotype, for quantitative genetic or other purposes, would ideally sample the many developmental and environmental factors that modulate it.

Quantitative Trait Loci (QTL) mapping is a powerful technique for dissecting the genetic basis of traits and species differences. Established tomato mapping populations between domesticated tomato (Solanum lycopersicum) and its more distant interfertile relatives typically follow a near isogenic line (NIL) design, such as the S. pennellii Introgression Line (IL) population, with a single wild introgression per line in an otherwise domesticated genetic background. Here, we report on a new advanced backcross QTL mapping resource for tomato, derived from a cross between the M82 tomato cultivar and S. pennellii. This so-called Backcrossed Inbred Line (BIL) population is comprised of a mix of BC2 and BC3 lines, with domesticated tomato as the recurrent parent. The BIL population is complementary to the existing S. pennellii IL population, with which it shares parents. Using the BILs, we mapped traits for leaf complexity, leaflet shape, and flowering time. We demonstrate the utility of the BILs for fine-mapping QTL, particularly QTL initially mapped in the ILs, by fine-mapping several QTL to single or few candidate genes. Moreover, we confirm the value of a backcrossed population with multiple introgressions per line, such as the BILs, for epistatic QTL mapping. Our work was further enabled by the development of our own statistical inference and visualization tools, namely a heterogeneous hidden Markov model for genotyping the lines, and by using state-of-the-art sparse regression techniques for QTL mapping.