Chicken domestication: an updated perspective based on mitochondrial genomes

Domestic chickens (Gallus gallus domesticus) fulfill various roles ranging from food and entertainment to religion and ornamentation. To survey its genetic diversity and trace the history of domestication, we investigated a total of 4938 mitochondrial DNA (mtDNA) fragments including 2843 previously published and 2095 de novo units from 2044 domestic chickens and 51 red junglefowl (Gallus gallus). To obtain the highest possible level of molecular resolution, 50 representative samples were further selected for total mtDNA genome sequencing. A fine-gained mtDNA phylogeny was investigated by defining haplogroups A–I and W–Z. Common haplogroups A–G were shared by domestic chickens and red junglefowl. Rare haplogroups H–I and W–Z were specific to domestic chickens and red junglefowl, respectively. We re-evaluated the global mtDNA profiles of chickens. The geographic distribution for each of major haplogroups was examined. Our results revealed new complexities of history in chicken domestication because in the phylogeny lineages from the red junglefowl were mingled with those of the domestic chickens. Several local domestication events in South Asia, Southwest China and Southeast Asia were identified. The assessment of chicken mtDNA data also facilitated our understanding about the Austronesian settlement in the Pacific.     

Variations in genome size between wild and domesticated lineages of fowls belonging to the Gallus gallus species

Efforts to elucidate the causes of biological differences between wild fowls and their domesticated relatives, the chicken, have to date mainly focused on the identification of single nucleotide mutations. Other types of genomic variations have however been demonstrated to be important in avian evolution and associated to variations in phenotype. They include several types of sequences duplicated in tandem that can vary in their repetition number.Here we report on genome size differences between the red jungle fowl and several domestic chicken breeds and selected lines. Sequences duplicated in tandem such as rDNA, telomere repeats, satellite DNA and segmental duplications were found to have been significantly re-shaped during domestication and subsequently by human-mediated selection. We discuss the extent to which changes in genome organization that occurred during domestication agree with the hypothesis that domesticated animal genomes have been shaped by evolutionary forces aiming to adapt them to anthropized environments.     

Domestication effects on foraging strategies in fowl

Birds of two different breeds differing in degree of domestication were studied to reveal any differences in foraging strategies between them. The breeds were wild-type birds (crossing between red jungle fowl (Gallus gallus) and Swedish bantam (Gallus gallus domesticus) and domestic birds (Swedish bantam), breeds representing an increasing level of domestication. Bantam birds have not been selected for any specific characteristics. The birds were allowed to forage in an experimental pen containing two separate food patches, which depleted as a function of being exploited, to see how well the different breeds were able to assess costs and benefits as the distance between patches were changed (short distance between patches compared to long distance between patches). Both breeds behaved in accordance with some general predictions of optimal foraging theory, i.e. moved between patches, left patches before these were empty and stayed for a shorter time in more depleted patches. Wild-type birds responded more than domestic birds to an increase of distance between patches, by spending longer average time in patch when there was a long distance between them compared to when there was a short distance. The wild-type birds adopted what seemed to be a more costly foraging strategy, moving more between patches than the domestic birds without ingesting more feed. During domestication, in the protected environment provided by man, individuals using less costly behavioural strategies may have gained increased fitness over those spending more energy on foraging. Although domestic birds still possessed the ability to respond adaptively to environmental conditions, the differences between the wild-type and the domestic breed might be a result of the reduction of the natural selection pressure which accompanies domestication.     

Evolutionary pets: offspring numbers reveal speciation process in domesticated chickens

Since Darwin, the nature of the relationship between evolution and domestication has been debated. Evolution offers different mechanisms of selection that lead to adaptation and may end in the origin of new species as defined by the biological species concept. Domestication has given rise to numerous breeds in almost every domesticated species, including chickens. At the same time, so-called artificial selection seems to exclude mechanisms of sexual selection by the animals themselves. We want to forward the question to the animal itself: With whom do you reproduce successfully? This study focused on the sexual behavior of the domestic chicken Gallus gallus f.dom., particularly the White Crested Polish breed. Experiments on mate choice and the observation of fertilization and hatching rates of mixed-breeding groups revealed breed-specific preferences. In breeding groups containing White Crested Polish and a comparative breed, more purebred chicks hatched than hybrids (number of eggs collected: 1059). Mating was possible in equal shares, but in relation to the number of eggs collected, purebred offspring (62.75%±7.10%, M±SE) hatched to a greater extend compared to hybrid offspring (28.75%±15.32%, M±SE). These data demonstrate that the mechanism of sexual selection is still present in domestic chicken breeds, which includes the alteration of gene frequencies typical for domestication and evolutionary speciation. Due to selection and mate choice we state that breeding in principle can generate new species. Therefore, we see domestication as an evolutionary process that integrates human interests of animal breeding with innate mate choice by the animal.     

Whole-genome resequencing reveals domestication and signatures of selection in Ujimqin,Sunit, and Wu Ranke Mongolian sheep breeds

ObjectiveThe current study aimed to perform whole-genome resequencing of Chinese indigenous Mongolian sheep breeds including Ujimqin, Sunit, and Wu Ranke sheep breeds (UJMQ, SNT, WRK) and deeply analyze genetic variation, population structure, domestication, and selection for domestication traits among these Mongolian sheep breeds.MethodsBlood samples were collected from a total of 60 individuals comprising 20 WRK, 20 UJMQ, and 20 SNT. For genome sequencing, about 1.5 μg of genomic DNA was used for library construction with an insert size of about 350 bp. Pair-end sequencing were performed on Illumina NovaSeq platform, with the read length of 150 bp at each end. We then investigated the domestication and signatures of selection in these sheep breeds.ResultsAccording to the population and demographic analyses, WRK and SNT populations were very similar, which were different from UJMQ populations. Genome wide association study identified 468 and 779 significant loci from SNT vs UJMQ, and UJMQ vs WRK, respectively. However, only 3 loci were identified from SNT vs WRK. Genomic comparison and selective sweep analysis among these sheep breeds suggested that genes associated with regulation of secretion, metabolic pathways including estrogen metabolism and amino acid metabolism, and neuron development have undergone strong selection during domestication.ConclusionOur findings will facilitate the understanding of Chinese indigenous Mongolian sheep breeds domestication and selection for complex traits and provide a valuable genomic resource for future studies of sheep and other domestic animal breeding.     

云南地方鸡种的遗传多样性及其与中国家鸡起源的关系

云南地方鸡种在生境、形态外貌、细胞遗传和血液蛋白等方面均表现出多样性,但mtDNA的遗传变异单一。云南地方鸡种有其独特的基因类型,由于所处的特殊地理位置和生态条件差异,形成了与我国其它鸡种相对独立的鸡种。从血液蛋白、细胞遗传和mtDNA等方面进一步证明红色原鸡为家鸡的祖先。云南可能是中国家鸡的起源中心之一。     

Accumulation of deleterious mutations in the domestic yak genome

Deleterious mutations play an important functional role, affecting trait phenotypes in ways that decrease the fitness of organisms. Estimating the frequency of occurrence and abundance has been a topic of much interest, especially in crops and livestock. The processes of domestication and breeding allow deleterious mutations to persist at high frequency, and identifying such deleterious mutations is particularly important for breed improvement. Here, we assessed genome‐wide patterns of deleterious variation in 59 domestic and 13 wild yaks using genome resequencing data. Based on the intersection of results given by three methods (provean , polyphen 2 and sift 4g ), we identified 3187 putative deleterious mutation sites affecting 2586 genes in domestic yaks and 2067 affecting 1701 genes in wild yaks. Multiple lines of evidence indicate a significant increase in the load of deleterious mutations in domesticated yaks compared to wild yaks. Private deleterious genes were found to be associated with the perception of smell and detection of chemical stimulus. We also identified 36 genes related to Mendelian genetic diseases involved in sensory perception, skeletal development and the nervous and immune systems. This study not only adds to the understanding of the genetic basis of yak domestication but also provides a rich catalog of variants that will facilitate future breeding‐related research on the yak genome and on other bovid species.     

Analysis of genetic diversity and phylogenetic relationships among red jungle fowls and Chinese domestic fowls

Genetic diversity and phylogenetic relationships among 568 individuals of two red jungle fowl subspe- cies (Gallus gallus spadiceus in China and Gallus gallus gallus in Thailand) and 14 Chinese domestic chicken breeds were evaluated with 29 microstaellite loci, the genetic variability within population and genetic differentiation among population were estimated, and then genetic diversity and phylogenetic relationships were analyzed among red jungle fowls and Chinese domestic fowls. A total of 286 alleles were detected in 16 population with 29 microsatellite markers and the average number of the alleles observed in 29 microsatellite loci was 9.86±6.36. The overall expected heterozygosity of all population was 0.6708±0.0251, and the number of population deviated from Hardy-Weinberg equilibrium per locus ranged from 0 to 7. In the whole population, the average of genetic differentiation among population, measured as FST value, was 16.7% (P<0.001), and all loci contributed significantly (P<0.001) to this differentiation. It can also be seen that the deficit of heterozygotes was very high (0.015) (P<0.01). Reynolds' distance values varied between 0.036 (Xiaoshan chicken-Luyuan chicken pair) and 0.330 (G. gallus gallus-Gushi chicken pair). The Nm value ranged from 0.533 (between G. gallus gallus and Gushi chicken) to 5.833 (between Xiaoshan chicken and Luyuan chicken). An unrooted consensus tree was constructed using the neighbour-joining method and the Reynolds' genetic distance. The heavy-body sized chicken breeds, Luyuan chicken, Xiaoshan chicken, Beijing Fatty chicken, Henan Game chicken, Huainan Partridge and Langshan chicken formed one branch, and it had a close genetic relationship between Xiaoshan chicken-Luyuan chicken pair and Chahua chicken-Tibetan chicken pair. Chahua chicken and Tibetan chicken had closer genetic relationship with these two subspecies of red jungle fowl than other domestic chicken breeds. G. gallus spadiceus showed closer phylogenetic relationship with Chinese domestic chicken breeds than G. gallus gallus. All 29 microstaellite loci in this study showed high levels of polymorphism and significant genetic differentiation was observed among two subspecies of red jungle fowl and 14 Chinese domestic chicken breeds. The evolutional dendrogram is as follows: evolutional breeds→primitive breeds (Chahua chicken and Tibetan)→red jungle fowl in China (G. gallus spadiceus)→red jungle fowl in Thailand (G. gallus gallus). The results supported the theory that the domestic fowls might originate from different subspecies of red jungle fowl and Chinese domestic fowls had independent origin.     

The origin and genetic diversity of Chinese native chicken breeds

The first 539 bases of mitochondrial DNA D-loop region of six Chinese native chicken breeds (Gallus gallus domesticus) were sequenced and compared to those of the red junglefowl (Gallus gallus), the gray junglefowl (Gallus sonneratii), the green junglefowl (Gallus varius) and Lafayette's junglefowl (Gallus lafayettei) reported in GenBank, and the phylogenetic trees for the chickens were constructed based on the D-loop sequences. The results showed that the four species of the genus Gallus had great differences among each other, the G. g. domesticus was closest to the red junglefowl in Thailand and its adjacent regions, suggesting the Chinese domestic fowl probably originated from the red junglefowl in these regions. The two subs pecies of Thailand, G. g. gallus and G. g. spadiceus, should belong to one subspecies because of their resemblance. In the case of native breeds, there existed a great difference between the egg breeds and general purpose breeds, which suggested different maternal origins of the two types.     

Domestic chickens defy Rensch's rule: sexual size dimorphism in chicken breeds

Sexual size dimorphism (SSD), i.e. the difference in sizes of males and females, is a key evolutionary feature that is related to ecology, behaviour and life histories of organisms. Although the basic patterns of SSD are well documented for several major taxa, the processes generating SSD are poorly understood. Domesticated animals offer excellent opportunities for testing predictions of functional explanations of SSD theory because domestic stocks were often selected by humans for particular desirable traits. Here, we analyse SSD in 139 breeds of domestic chickens Gallus gallus domesticus and compare them to their wild relatives (pheasants, partridges and grouse; Phasianidae, 53 species). SSD was male-biased in all chicken breeds, because males were 21.5 ± 0.55% (mean ± SE) heavier than females. The extent of SSD did not differ among breed categories (cock fighting, ornamental and breeds selected for egg and meat production). SSD of chicken breeds was not different from wild pheasants and allies (23.5 ± 3.43%), although the wild ancestor of chickens, the red jungle fowl G. gallus, had more extreme SSD (male 68.8% heavier) than any domesticated breed. Male mass and female mass exhibited positive allometry among pheasants and allies, consistently with the Rensch's rule reported from various taxa. However, body mass scaled isometrically across chicken breeds. The latter results suggest that sex-specific selection on males vs. females is necessary to generate positive allometry, i.e. the Rensch's rule, in wild populations.     

Structural Variants Selected during Yak Domestication Inferred from Long-Read Whole-Genome Sequencing

Structural variants (SVs) represent an important genetic resource for both natural and artificial selection. Here we present a chromosome-scale reference genome for domestic yak (Bos grunniens) that has longer contigs and scaffolds (N50 44.72 and 114.39 Mb, respectively) than reported for any other ruminant genome. We further obtained long-read resequencing data for 6 wild and 23 domestic yaks and constructed a genetic SV map of 372,220 SVs that covers the geographic range of the yaks. The majority of the SVs contains repetitive sequences and several are in or near genes. By comparing SVs in domestic and wild yaks, we identified genes that are predominantly related to the nervous system, behavior, immunity, and reproduction and may have been targeted by artificial selection during yak domestication. These findings provide new insights in the domestication of animals living at high altitude and highlight the importance of SVs in animal domestication.     

鸡SNP多样性的比较研究与群体有效规模的估算

以红色原鸡(Red Jungle Fowl, RJF)、丝羽乌骨鸡(Taihe silk Chicken, TS)、隐性白洛克鸡(White Recessive Rock, WRR)为资源群, 在鸡一号染色体的Contig.060226.1上选取了一个200 kb的区域, 比较研究了3个群体的SNP (single nucleotide polymorphism)多样性、估算了鸡的初始群体有效规模大小(effective size of population, Ne)。红色原鸡、丝羽乌骨鸡、隐性白洛克鸡3个群体的平均杂合度分别为0.28533±0.03475、0.32926±0.03919、0.30168±0.04038。显著性检验差异不显著(P=0.2368>0.05)。根据Latter 和Nei的方法对鸡的群体有效规模进行了估算, 鸡的初始群体有效规模大小为20 000～150 000。鸡在驯养的早期阶段经历过严厉的瓶颈效应, 但瓶颈效应对鸡各品种SNP的多样性并未产生显著影响。笔者认为, 鸡在驯养的早期阶段群体有效规模足够大, 品种分化过程中群体迅速扩张, 品种间的广泛杂交(特别是和红色原鸡之间)以及鸡基因组的高重组率等因素是导致家鸡和原鸡以及各家鸡品种间SNP多样性没有显著差别的重要原因。     

Globally dispersed Y chromosomal haplotypes in wild and domestic sheep

To date, investigations of genetic diversity and the origins of domestication in sheep have utilised autosomal microsatellites and variation in the mitochondrial genome. We present the first analysis of both domestic and wild sheep using genetic markers residing on the ovine Y chromosome. Analysis of a single nucleotide polymorphism (oY1) in the SRY promoter region revealed that allele A-oY1 was present in all wild bighorn sheep (Ovis canadensis), two subspecies of thinhorn sheep (Ovis dalli), European Mouflon (Ovis musimon) and the Barbary (Ammontragis lervia). A-oY1 also had the highest frequency (71.4%) within 458 domestic sheep drawn from 65 breeds sampled from Africa, Asia, Australia, the Caribbean, Europe, the Middle East and Central Asia. Sequence analysis of a second locus, microsatellite SRYM18, revealed a compound repeat array displaying fixed differences, which identified bighorn and thinhorn sheep as distinct from the European Mouflon and domestic animals. Combined genotypic data identified 11 male-specific haplotypes that represented at least two separate lineages. Investigation of the geographical distribution of each haplotype revealed that one (H6) was both very common and widespread in the global sample of domestic breeds. The remaining haplotypes each displayed more restricted and informative distributions. For example, H5 was likely founded following the domestication of European breeds and was used to trace the recent transportation of animals to both the Caribbean and Australia. A high rate of Y chromosomal dispersal appears to have taken place during the development of domestic sheep as only 12.9% of the total observed variation was partitioned between major geographical regions.     

Genetic structure of a wide-spectrum chicken gene pool

The genetic structure of 65 chicken populations was studied using 29 simple sequence repeat loci. Six main clusters which corresponded to geographical origins and histories were identified: Brown Egg Layers; predominantly Broilers; native Chinese breeds or breeds with recent Asian origin; predominantly breeds of European derivation; a small cluster containing populations with no common history and populations that had breeding history with White Leghorn. Another group of populations that shared their genome with several clusters was defined as 'Multi-clusters'. Gallus gallus gallus (Multi-clusters), one of the subspecies of the Red Jungle Fowl, which was previously suggested to be one of the ancestors of the domesticated chicken, has almost no shared loci with European and White Egg layer populations. In a further sub-clustering of the populations, discrimination between all the 65 populations was possible, and relationships between each were suggested. The genetic variation between populations was found to account for about 34% of the total genetic variation, 11% of the variation being between clusters and 23% being between populations within clusters. The suggested clusters may assist in future studies of genetic aspects of the chicken gene pool.     

Molecular evidence for hybridization of species in the genus Gallus except for Gallus varius

A phylogenetic tree for fowl including chicken in the genus Gallus and based on mitochondrial D-loop analysis further supports the hypothesis developed from morphology and progeny production that red junglefowl (RJF) is the direct ancestor of the chicken. The phylogenetic positions of the chicken and the other fowl species in the genus Gallus are of great importance when considering maintenance and improvement of chicken breeds through introgression of genetic variation from wild-type genomes. However, because the phylogenetic analysis based on the DNA sequences is not sufficient to conclude the phylogenetic positions of the fowls in the genus, in the present study, we have determined sequences of whole mitochondrial DNA (mtDNA) and two segments of the nuclear genome (intron 9 of ornithine carbamoyltransferase, and four chicken repeat 1 elements) for the species in the genus Gallus. The phylogenetic analyses based on mtDNA sequences revealed that two grey junglefowls (GyJF) were clustered in a clade with RJFs and chicken, and that one GyJF was located in a remote position close to Ceylon junglefowl (CJF). The analyses based on the nuclear sequences revealed that alleles of GyJFs were alternatively clustered with those of CJF and with those of RJFs and chicken. Alternative clustering of RJF and chicken alleles were also observed. These findings taken together strongly indicate that inter-species hybridizations have occurred between GyJF and RJF/chicken and between GyJF and CJF.     

The distribution of endogenous chicken retrovirus sequences in the DNA of galliform birds does not coincide with avian phylogenetic relationships.

The chicken is a domesticated form of Red Jungle-fowl (Gallus gallus), which belongs to the Pheasant family (Phasianidae) within the order Galliformes. Domestic chickens carry the genome of the endogenous retrovirus RAV-O as DNA sequences integrated into host chromosomes transmitted through the germ line. We have examined the presence and distribution of RAV-O-related sequences in the DNA of Red Junglefowl and other closely related species of Junglefowl, as well as more distantly related Pheasants and Quail. DNA sequences homologous to RAV-O were analyzed by molecular hybridization in liquid and after electrophoresis of restriction endonuclease fragments. The presence of RAV-O-related sequences in avian DNA does not correlate with phylogenetic relationships. Under stringent conditions of hybridization in liquid, DNA sequences homologous to RAV-O cDNA were detected at high levels (greater than 80% homology( only in the genomes of the domestic chicken and its phylogenetic ancestor, the Red Junglefowl (Gallus gallus). The DNA of two other species of Gallus (G. sonnerati, Sonnerat's Junglefowl and G. varius, Green Junglefowl), of Ring-necked Pheasant and of Japanese Quail contained sequences with less than 10% homology to RAV-O cDNA. Under conditions permitting mismatching, however, Ring-necked Pheasant DNA hybridized up to 50% of the RAV-O cDNA, and Quail DNA 24%, whereas the extent of hybridization to Sonnerat's and Green Junglefowl DNA was not markedly increased. Analysis of restriction enzyme digests revealed several distinct fragments of DNA hybridizing to chick retrovirus cDNA in both Red Junglefowl and domestic chicken, and multiple fragments in DNA from two species of Phasianus. No fragments with sequences related to chicken retroviruses were found, however, in digests of DNA prepared from Sonnerat's, Ceylonese and Green Junglefowl, from two other Pheasant genera (Chrysolophus and Lophura), or from one Quail genus (Coturnix). Thus the DNA of three Junglefowl species closely related to Gallus gallus lacked RAV-O sequences while the DNA of more distantly related Phasianus species showed significant homology. These results show that RAV-O-related sequences have not diverged together with the normal host genes during the evolution of the Phasianidae. Although RAV-O sequences are endogenous in all domestic chickens and Red Junglefowl studied thus far, it appears that the RAV-O genome has been introduced relatively recently into the germ line of Gallus gallus, following speciation but before domestication, and independently of the related sequences found in members of the genus Phasianus.     

Application of functional genomic information to develop efficient EST-SSRs for the chicken (Gallus gallus)

Many years of domestication and breeding have given rise to the wide range of chicken breeds that exist today; however, an increasing number of local chicken breeds are under threat of extinction. A comprehensive characterization of chicken markers (especially type I markers) is needed to monitor and conserve genetic diversity in this species. The explosion of genomics and functional genomics information in recent years has opened new possibilities for the generation of molecular markers. We analyzed a large number of expressed sequence tags (ESTs) to test the possibility of using EST-derived microsatellite markers for investigating the Gallus gallus genome. Chromosomal locations for the majority of these SSRs were predicted. Of the 31,576 unigenes assembled from the 544,150 redundant EST sequences, 1757 SSR markers were discovered on 1544 ESTs, using the SSRLocator software, with an average density of 28.7 kb per SSR. The dimer motifs were the most frequent (46.38%), followed by trimeric (38.58%), tetrameric (10.19%), pentameric (4.5%), and hexameric (<1%) markers. Different from the case for cattle and sheep, AT/TA was the most abundant dimeric repeat, accounting for 41.71% of all dimeric repeats in the chicken ESTs. The EST-SSR distribution was not uniform among the chromosomes; the majority of the EST-SSRs were located on chromosomes GGA2 and GGA10. We found that most of the EST-SSRs are involved in positive regulation of cellular and metabolic processes. This is the first time that EST sequences have been mined to find chicken microsatellites. On average, 3.8% of the G. gallus UniGene sequences could be exploited for development of EST-SSRs, indicating a good source for molecular markers as well as for functional genome analysis.     

Intronic deletions that disrupt mRNA splicing of the tva receptor gene result in decreased susceptibility to infection by avian sarcoma and leukosis virus subgroup A

The group of closely related avian sarcoma and leukosis viruses (ASLVs) evolved from a common ancestor into multiple subgroups, A to J, with differential host range among galliform species and chicken lines. These subgroups differ in variable parts of their envelope glycoproteins, the major determinants of virus interaction with specific receptor molecules. Three genetic loci, tva, tvb, and tvc, code for single membrane-spanning receptors from diverse protein families that confer susceptibility to the ASLV subgroups. The host range expansion of the ancestral virus might have been driven by gradual evolution of resistance in host cells, and the resistance alleles in all three receptor loci have been identified. Here, we characterized two alleles of the tva receptor gene with similar intronic deletions comprising the deduced branch-point signal within the first intron and leading to inefficient splicing of tva mRNA. As a result, we observed decreased susceptibility to subgroup A ASLV in vitro and in vivo. These alleles were independently found in a close-bred line of domestic chicken and Indian red jungle fowl (Gallus gallus murghi), suggesting that their prevalence might be much wider in outbred chicken breeds. We identified defective splicing to be a mechanism of resistance to ASLV and conclude that such a type of mutation could play an important role in virus-host coevolution.     

Genomic landscape of parallel domestication of upland rice and its implications

Parallel domestication has been widely acknowledged but itsgenetic basis remains largely unclear. As an important rice ecotype, upland rice was assumedly domesticated multiple times in two rice subspecies (Indica and Japonica) and provides a feasible system to explore the genetic basis of parallel domestication. To uncover the genome‐wide pattern of genetic differentiation between upland and lowland rice and explore the parallelism of genetic changes during upland rice domestication, we obtained whole‐genome sequences of 95 rice landraces and yielded genome‐wide expression data for five tissues of representative accessions of upland and lowland rice. Our phylogenetic analyses confirmed multiple domestications of the upland ecotype in two rice subspecies. Genomic scans based on resequencing data identified substantial differentiation between the upland and lowland ecotypes with 11.4% and 14.8% of the genome diverged between the two ecotypes in Indica and Japonica, respectively. Further genome‐wide gene expression analyses found that 30% of effectively expressed genes were significantly differentiated between two ecotypes, indicating the importance of regulation changes in the domestication of upland rice. Importantly, we found that only 1.8% of differentiated genomes and 1.6% of differentially expressed genes were shared by upland Indica and upland Japonica, suggestive of largely unparallel genetic alterations during upland rice domestication. These findings not only provide new insights into the genetic basis of parallel domestication at the genome scale but could also facilitate geneticimprovement and breeding of rice and crops in general.