Each locus will have an allelic form allele. The complete set of alleles at all loci of interest in an individual is its genotype. Typically, when writing out a genotype, only the alleles at the locus loci of interest are considered — all the others are present and assumed to be wild type. The visible or detectable effect of these alleles on the structure or function of that individual is called its phenotype — what it looks like. The phenotype studied in any particular genetic experiment may range from simple, visible traits such as hair color, to more complex phenotypes including disease susceptibility or behavior.
If two alleles are present in an individual, then various interactions between them may influence their expression in the phenotype. We have already said that one allele as a homozygote produces purple flowers, while the other allele as a homozygote produces white flowers see Figures 1. But what about an individual that has one purple allele and one white allele; what is the phenotype of an individual whose genotype is heterozygous?
This can only be determined by experimental observation. We know from observation that individuals heterozygous for the purple and white alleles of the flower color gene have purple flowers. Thus, the allele associated with purple color is therefore said to be dominant to the allele that produces the white color.
The white allele, whose phenotype is masked by the purple allele in a heterozygote, is recessive to the purple allele. To represent this relationship, often, a dominant allele will be represented by a capital letter e. A while a recessive allele will be represented in lower case e. Improving persistence is therefore, a major breeding goal in most breeding programs Taylor, ; Boller et al. Red clover is an outbreeding species with a gametophytic self-incompatibility system Taylor, , and thus there is a considerable amount of genetic variation within cultivars, which are usually synthetic populations with a large number of parents.
Red clover has a genome size of Mb Sato et al. This facilitates particularly the identification of genomic regions potentially under selection. Persistence is a complex trait controlled by many different genetic and environmental factors.
It may be defined purely as survival over years, or alternatively, as maintenance of annual yield over years. These two measures can be correlated e. Biotic and abiotic stresses such as fungal pathogens, nematodes and insect herbivores, abiotic stresses, and competition from grasses under high N fertilization levels, are factors that can cause reduced persistence of red clover Lager and Gerhardson, ; Abberton and Marshall, ; Taylor, ; Boller et al.
Cutting frequency has been found to be of lesser importance Coulman and Kielly, ; Wiersma et al. Positive correlations have been found between persistence and stem height or leaf size Herrmann et al. The ability to regrow after repeated defoliation is likely to be related to presence of leaves low in the canopy, level of root energy reserves and number of crown buds, as indicated for lucerne Brummer and Bouton, , Little is known about the genetic control of persistence, but it is likely that interaction with environmental factors are important.
To our knowledge only one report on QTLs for persistence Herrmann et al. Only a fraction of the sown plants will survive the first years due to competition and stress. However, the initial survival rate is likely higher in species mixtures because intraspecific competition is usually higher than interspecific competition. Natural selection can guide breeding Henry and Nevo, , particularly when it comes to a trait like the survival component of persistence.
Selection of high persistence in breeding programs is usually done by selecting plants that have survived under field conditions for around 3 years. Such survivor populations of both red and white clover have been found to have experienced a shift in the genetic composition of the populations relative to the original populations, measured either with molecular markers or with phenotyping of offspring Annicchiarico and Piano, ; Collins et al.
In this paper, we utilized survivor populations to investigate whether non-random selection could be detected within one generation of red clover growing in the field for 2.
We first aimed to identify loci controlling persistence by detecting loci with significantly altered allele frequencies in survivor populations compared to the originally sown population study 1. We based our analyses on single nucleotide polymorphism SNP data obtained from genotyping by sequencing GBS of individuals from the original population and survivor populations. We then tested whether reliable allele frequencies could be obtained by sequencing pools of individual DNA samples rather than sequencing the individual samples themselves study 2.
Finally, we used GBS-derived SNP data from pools of leaf samples to investigate whether different loci had been selected in red clover survivors grown in pure stands as compared to red clover survivors grown in species mixtures study 3. The plot size was 7. With an approximate thousand seed weight of diploid red clover of 2 g, this equals approximately or seeds per Ps plot, and or seeds per Ms plot.
Plots had been harvested either 3 or 5 times a year 3H and 5H. Leaf blades were sampled from survivor populations i. In , DNA was extracted from leaves of 48 or 47 survivor plants randomly selected from Ps survivor populations sown at high seeding rate and harvested 3 or 5 times a year two plots from each harvesting regime, sample set 1, Table 1 , and from leaf samples of 88 individuals of the original population seeded in the greenhouse sample set 2 , using DNeasy 96 Plant Kit Qiagen.
GBS libraries were made for each of the individuals. In order to test how well allele frequencies can be estimated from DNA pools, equal amounts of DNA from the 88 individuals of the original population were pooled and distributed among seven tubes from which 7 replicate GBS libraries were made sample set 3. Leaves were sampled from four red clover Ps survivor populations, and four Ms survivor populations, all which had been harvested 3 times a year. Two plots of each stand type had been sown at high seeding rate H and two at low seeding rate L.
From the Ps H populations, three replicate samples, each consisting of one leaf from each of random individuals, were sampled in order to evaluate the reproducibility of the sampling. For the libraries made from individuals in sample set 1 and 2 Table 1 , a minimum of ten reads in total, and, in the case of heterozygotes, a minimum of two reads for each of the alleles, was required to maintain the genotype for each genotype and SNP combination. For visualization of population structure the genotype of each SNP and all individuals were used as input data.
In order to identify SNP loci that had been under selection two different methods were used. Thirdly, SNPs with significant F ST in all four survivor populations relative to the original population, were identified. Only these SNPs were regarded as having different allele frequency in survivor populations as compared to the original population.
The 64 bp tag sequences containing significant SNPs were blasted against the red clover draft genome De Vega et al. Single nucleotide polymorphism outliers were also detected with BayeScan v 2. The allele frequency dataset described above was converted to allele numbers using the number of haploid genomes that had been genotyped for each population. Population-specific F ST -values are based on the comparison between each population and the pool of all populations in the analysis.
Outlier SNPs are identified as those, where the SNP-specific component is necessary to explain the observed variation. We ran the analysis with five defined populations four survivor populations and the original population and three defined populations data from 3H populations pooled, data from 5H populations pooled, original population. For the seven libraries made from pools of individual DNA samples sample set 3, Table 1 , the sequence reads were filtered for minimum 10 and maximum reads per SNP and replicate library.
Reads were subsequently pooled across libraries, and allele frequencies were calculated based on number of reads. For comparison of allele frequencies obtained from GBS of individuals sample set 2 vs. We discarded those that mapped to scaffolds not yet assigned a chromosomal locations, leaving SNPs for analysis.
Allele frequencies were estimated on the basis of the number of reads. Differentiation between populations and replicate samples was analyzed with PCA as in experiment 1. In order to identify SNPs potentially differentially selected in plots receiving different treatments, we used the same two methods as in study 1.
For the simple F ST -based method, pairwise F ST values between each of the four Ps populations and the average of the Ms populations, and between each of the four Ms populations and the average of the Ps populations, were calculated for each SNP. For the two plots with replicate samples, the average allele frequencies for each plot were used. Allele frequency differences were tested for significance as described in study 1. For a SNP locus to be identified as having been under differential selection, it was required that all eight F ST values were significant.
Fishers exact test was not performed due to the lack of individual genotypes. BayeScan compares each population with all the other populations in the analysis. An identified SNP outlier may therefore not necessarily vary consistently between, e. We also ran a BayeScan analysis in which the data from each of the eight populations were combined in two main groups, Ps and Ms.
The two populations from the 5H harvesting regime were more similar to each other than the two populations from the 3H harvesting regime. These results suggest that the survivor populations had diverged from the original population in different directions, and that most of the observed difference in allele frequencies was random. Figure 1. Genetic differentiation between the original population Orig, 88 individuals and four survivor populations sampled from four plots 47—48 individuals per population in a field experiment 2.
The survivor populations had been harvested three 3H or five 5H times a year, and there were two replicate plots of each harvesting regime. The three first principal components explaining the variation in allele frequencies are shown. The same weak structure observed in the original population, with one major and one minor subpopulation, was observed in each of four survivor populations Figure 2.
The proportion of individuals in the minor cluster were 0. Figure 2. Population structure revealed by PCA of SNP genotype data for 88 individuals from the original population together with 47 or 48 individuals from each of the four survivor populations in study 1. The SNP loci were first screened for significant shifts in allele frequency between the original population and survivor populations with a simple test using F ST values and a chi square test combined with FDR.
When testing individual survivor populations separately, or when combining all survivors in one common population, significant FDR values could not be obtained; they were in the range of 0. Thus, the number of SNP loci identified was only about twice of what could be expected from chance alone. However, it is very unlikely that a locus has a shift in allele frequency due to chance alone in all four survivor populations.
The average shifts in allele frequency of the 27 SNPs across the four survivor populations ranged from 0. For comparison, the F ST value averaged across survivor populations and all loci was 0. Sequence tags of 20 of the 27 SNPs could be mapped onto the red clover genome. Fourteen SNPs mapped to chromosomes between two and four SNPs on each of chromosome 1, 2, 3, 4, and 7 Table 2 and Figure 3 , and six mapped to scaffolds not yet assigned to chromosomes Supplementary Table S2.
Table 2. Sequence tags containing SNPs with a significant shift in allele frequency in four red clover pure stand survivor populations relative to the originally sown population study 1 , identified using a simple F ST -based method. Figure 3. Map positions of chromosomal SNPs found to have been under selection. Black bold; selected in four survivor populations pure stand, sown at high density as compared to the original, sown population study 1 , red; differentially selected in survival populations from pure stand as compared to survival populations from species mixtures study 3 , green; differentially selected in populations as compared to populations sown at low seeding density study 3 , pink italic, selection affected by both stand type and seeding density study 3.
We also attempted to identify SNP loci that had been under specific selection in either of the two harvesting regimes by looking for significant allele frequency differences between the 3H and the 5H survivor populations, using the simple F ST -based method. No such SNP loci could be detected, possibly due to the fact that there were only two replicate plots of each harvesting regime, and thus less power in the test.
A comparison was made between the allele frequencies obtained by genotyping individuals from the original population, with the allele frequency estimates obtained when sequencing a pool of equal amounts of DNA from each individual sample set 2 and 3. The best accuracy was obtained between and reads. Expanding the range to — reads resulted in a much higher number of SNPs , an average deviation of 0. Table 3. Average deviation in allele frequency estimates obtained when genotyping pooled samples consisting of equal amounts of 88 individuals as compared to genotyping the individuals separately study 1.
Figure 4. Comparison of the allele frequencies obtained when genotyping 88 individuals separately x -axis with those obtained when genotyping pools of equal amount of DNA from each individual y -axis in study 2. Seven GBS libraries were made from the pooled sample and the reads from these were combined prior to calculation of allele frequencies. Study 3 served both to investigate the possibility of pooling individual leaf samples prior to DNA extraction and GBS, and to investigate the possible differential selection that had occurred in survivor populations as a result of different stand types and seeding densities.
A comparison was made between allele frequencies obtained from genotyping of DNA extracted from three replicate pools of leaf tissue from each of two Ps H populations, each pool consisting of leaves. The average pairwise deviation in allele frequency between replicates was 0. Principal component analysis PCA Figure 5A revealed that the random variation between replicate samples from the same population was at least as large as the variation between samples from different populations.
When allele frequencies were averaged across the three replicates this random variation appeared to be reduced Figure 5B , indicating that a large part of the variation between samples was due to random variation which was reduced with the averaging of the three replicates. Figure 5. Genetic differentiation between survivor populations sampled from four pure stand plots Ps, red clover only and four mixed stand plots Ms, red clover growing in mixture with white clover, perennial ryegrass and tall fescue , sown at high H or low L seeding density study 3.
MAF for Ps H populations was the average of the three replicate samples. The largest differences were between populations belonging to different stand types. An allele is a variant form of a gene. Some genes have a variety of different forms, which are located at the same position, or genetic locus, on a chromosome. Humans are called diploid organisms because they have two alleles at each genetic locus, with one allele inherited from each parent. Each pair of alleles represents the genotype of a specific gene.
Genotypes are described as homozygous if there are two identical alleles at a particular locus and as heterozygous if the two alleles differ.
Alleles contribute to the organism's phenotype, which is the outward appearance of the organism.
0コメント