Correcting Estimators of θ and Tajima's D for Ascertainment Biases Caused by the Single-Nucleotide Polymorphism Discovery Process

Most single-nucleotide polymorphism (SNP) data suffer from an ascertainment bias caused by the process of SNP discovery followed by SNP genotyping. The final genotyped data are biased toward an excess of common alleles compared to directly sequenced data, making standard genetic methods of analysis inapplicable to this type of data. We here derive corrected estimators of the fundamental population genetic parameter θ = 4N_eμ (N_e, effective population size; μ, mutation rate) on the basis of the average number of pairwise differences and on the basis of the number of segregating sites. We also derive the variances and covariances of these estimators and provide a corrected version of Tajima's D statistic. We reanalyze a human genomewide SNP data set and find substantial differences in the results with or without ascertainment bias correction.     

Mapping and diagnostic marker development for Soil-borne cereal mosaic virus resistance in bread wheat

Monogenically-inherited resistance to Soil-borne cereal mosaic virus (SBCMV) in hexaploid bread wheat cultivars ‘Tremie’ and ‘Claire’ was mapped on chromosome 5D. The two closest flanking markers identified in the Claire-derived mapping population, Xgwm469-5D and E37M49, are linked to the resistance locus at distances of 1 and 9 cm, respectively. Xgwm469-5D co-segregated with the SBCMV resistance in the Tremie-derived population and with the recently identified Sbm1 locus in the cv. Cadenza. This suggested that Tremie and Claire carry a resistance gene allelic to Sbm1, or one closely linked to it. The diagnostic value of Xgwm469-5D was assessed using a collection of SBCMV resistant and susceptible cultivars. Importantly, all susceptible genotypes carried a null allele of Xgwm469-5D, whereas resistant genotypes presumably related to either Claire and Tremie or Cadenza revealed a 152 or 154 bp allele of Xgwm469-5D, respectively. Therefore, Xgwm469-5D is well suited for marker assisted selection for SBCMV resistance.     

Ancient Origin of the Null Allele se 428 of the Human ABO-Secretor Locus (FUT2)

In human populations, a null allele having several nucleotide differences from the wild-type allele is segregating at the FUT2 locus (the ABO-Secretor locus) encoding α(1,2)fucosyltransferase. To estimate the age of the most recent common ancestor (MRCA) of these two alleles, we sequenced FUT2 homologues from chimpanzee, gorilla, orangutan, and green monkey. Since we did not detect acceleration or any heterogeneity in the substitution rate at this locus among these species, the age of the MRCA was estimated to be around 3 MYA, assuming the divergence time of human and chimpanzee to be 5 MYA. We developed a simple test to examine whether or not the old age of the MRCA of the FUT2 is consistent with that expected for two divergent neutral alleles sampled from a random mating population. An application of the test to the data at FUT2 indicated that the age of the MRCA is too old to be explained by the simple neutral assumptions, although our test depends on accurate estimation of the divergence time of human and chimpanzee in units of twice the human population size. Various possibilities including balancing selection are discussed to explain this old age of the MRCA. Received: 9 May 1999 / Accepted: 20 September 1999     

Analysis of phenotypic and microsatellite-based diversity of maize landraces in India,especially from the North East Himalayan region

The maize landraces in the North East Himalayan (NEH) region in India, especially in the Sikkim state, are morphologically highly diverse. The present study provides details of phenotypic and molecular characterization of a set of 48 selected maize landrace accessions, including the ‘Sikkim Primitives’ which have a unique habit of prolificacy (5–9 ears on a single stalk). Multi-location phenotypic evaluation of these 48 accessions revealed significant genetic variability for grain yield and its components, leading to identification of several promising accessions. Cluster analysis and PCA using nine morpho-agronomic characters clearly separated ‘Sikkim Primitives’ from the rest of the accessions. PCA revealed two principal components describing 90% of the total variation, with hundred kernel weight, ear length, ear diameter, number of kernels per ear and flowering behaviour forming the most discriminatory traits. The accessions were genotyped using 42 microsatellite or simple sequence repeat (SSR) markers using a ‘population bulk DNA fingerprinting strategy’, with allele resolution using an automated DNA Sequencer. The study revealed a high mean number of alleles per SSR locus (13.0) and high Polymorphism Information Content (PIC) value of 0.60. The analysis also led to identification of 163 private/unique alleles, differentiating 44 out of 48 accessions. Six highly frequent SSR alleles were detected at different loci (phi014, phi062, phi090, umc1266, umc1367 and umc2250) with individual frequencies ≥0.75. Some of these SSR loci were reported to tag specific genes/QTL for some important traits, indicating that chromosomal regions harboring these SSR alleles were not selectively neutral. Cluster analysis using Rogers’ genetic distance also revealed distinct genetic identity of the ‘Sikkim Primitives’ from the rest of the accessions in India, including Sikkim. Mantel’s test revealed significant and positive correlation between the phenotypic and molecular genetic dissimilarity matrices. The study was the first to portray the patterns of phenotypic and molecular diversity in the maize landraces from the NEH region in India.     

Sequences of the cDNAs and genomic DNAs encoding the S1, S7, S8, and Sf alleles from almond, Prunus dulcis

Partial genomic and cDNA sequences of the RNase alleles S1, S7, S8 and Sf were obtained from Prunus dulcis cvs ’Ne Plus Ultra’, ’Ferragnes’ and ’Nonpareil’ 15–1, and IRTA Selection 12–2. Total DNA was extracted from leaves, and cDNA was prepared from total RNA extracted from styles. The partial cDNA sequences of the S1 allele from ’Ferragnes’, and the S7 and S8 alleles from ’Nonpareil’ 15–1, matched those reported in the literature for the alleles Sb, Sc and Sd respectively. The sequences of the S1, S7, S8 and Sf alleles found in genomic DNA contained introns of 562, 1,530, 2,208 and 689 bp respectively. The exon/intron splice junction sites of all alleles followed the GT/AG consensus sequence rule, and the sequences were found to be highly conserved. Received: 18 October 2000 / Accepted: 8 March 2001     

Associative overdominance: Evaluating the effects of inbreeding and linkage disequilibrium

Expressions are obtained for the expected phenotypic values of homozygous and heterozygous genotypes for a neutral marker locus linked to a locus segregating for a recessive deleterious gene. The phenotypic values are functions of the allele frequencies at the marker locus, the inbreeding coefficient and the degree of association of the deleterious gene with the marker alleles. The analysis is extended to more than two alleles at the marker locus. Either linkage disequilibrium or inbreeding alone can produce an apparent superiority of heterozygotes for the marker locus (unless specified otherwise, the terms ‘homozygote’ and ‘heterozygote’ will refer to the marker locus). The effect of linkage disequilibrium on the difference between the heterozygote and homozygote values can be positive (associative overdominance) or negative (associative underdominance), depending on the frequencies of the marker alleles and the degree of their association with the deleterious gene. Inbreeding has always a positive effect. In general, the expected value of a homozygote is a positive function of its allele frequency. When the various homozygous genotypes are combined into one class and the various heterozygous genotypes into another, the phenotypic difference of the two classes is a function of the evenness of the allelic frequency distribution. Inbreeding is a more likely explanation of associative overdominance if the frequency of the deleterious gene is low, but its effect on the character high. Conversely, linkage disequilibrium is more likely if the frequency is high and the effect low. The degrees of association between marker alleles and the deleterious gene can, in principle, be estimated from the observed phenotypic scores and used to calculate expected multi-locus genotype scores. This could provide the basis for statistical tests of the associative overdominance hypothesis as an explanation of observed correlations between multi-locus heterozygosity and phenotypic traits.     

Allozyme Variation and Phylogenetic Relationships in Picea jezoensis (Pinaceae) Populations of the Russian Far East

Genetic variation and differentiation of 12 populations of Picea jezoensis from the Russian Far East were studied using 20 allozyme loci. The mean number of alleles per locus was 2.63, the percent of polymorphic loci was 88.1%, the observed heterozygosity was 0.181, and the mean value of expected heterozygosity amounted to 0.189. The values of expected heterozygosity of the northern and central mainland populations were higher than in the southern part of the natural range. A significant bias of Hardy–Weinberg heterozygosity to equilibrium heterozygosity (H_eq) suggests that most of the mainland populations have recently experienced a severe expansion in population size while populations from Kamchatka Peninsula have undergone a reduction in population size. Unbiased Nei’s genetic distance values were low within and between the mainland and Sakhalin Island populations (D_N=0.008). The largest values (D_N=0.063) were found between the mainland/Sakhalin and Kamchatka Peninsula populations. Based on genetic distance, P. jezoensis and P. kamtschatkensis could be considered as distinct taxa, but P. ajanensis, P. microsperma, and P. komarovii do not warrant taxonomic recognition.     

Using a limited mapping strategy to identify major QTLs for resistance to grapevine powdery mildew (<Emphasis Type="Italic">Erysiphe necator</Emphasis>) and their use in marker-assisted breeding

A limited genetic mapping strategy based on simple sequence repeat (SSR) marker data was used with five grape populations segregating for powdery mildew (Erysiphe necator) resistance in an effort to develop genetic markers from multiple sources and enable the pyramiding of resistance loci. Three populations derived their resistance from Muscadinia rotundifolia ‘Magnolia’. The first population (06708) had 97 progeny and was screened with 137 SSR markers from seven chromosomes (4, 7, 9, 12, 13, 15, and 18) that have been reported to be associated with powdery or downy mildew resistance. A genetic map was constructed using the pseudo-testcross strategy and QTL analysis was carried out. Only markers from chromosome 13 and 18 were mapped in the second (04327) and third (06712) populations, which had 47 and 80 progeny, respectively. Significant QTLs for powdery mildew resistance with overlapping genomic regions were identified for different tissue types (leaf, stem, rachis, and berry) on chromosome 18, which distinguishes the resistance in ‘Magnolia’ from that present in other accessions of M. rotundifolia and controlled by the Run1 gene on chromosome 12. The ‘Magnolia’ resistance locus was termed as Run2.1. Powdery mildew resistance was also mapped in a fourth population (08391), which had 255 progeny and resistance from M. rotundifolia ‘Trayshed’. A locus accounting for 50% of the phenotypic variation mapped to chromosome 18 and was named Run2.2. This locus overlapped the region found in the ‘Magnolia’-based populations, but the allele sizes of the flanking markers were different. ‘Trayshed’ and ‘Magnolia’ shared at least one allele for 68% of the tested markers, but alleles of the other 32% of the markers were not shared indicating that the two M. rotundifolia selections were very different. The last population, 08306 with 42 progeny, derived its resistance from a selection Vitis romanetii C166-043. Genetic mapping discovered a major powdery mildew resistance locus termed Ren4 on chromosome 18, which explained 70% of the phenotypic variation in the same region of chromosome 18 found in the two M. rotundifolia resistant accessions. The mapping results indicate that powdery mildew resistance genes from different backgrounds reside on chromosome 18, and that genetic markers can be used as a powerful tool to pyramid these loci and other powdery mildew resistance loci into a single line.     

Amounts and patterns of nucleotide variation within and between two Japanese conifers, sugi (Cryptomeria japonica) and hinoki (Chamaecyparis obtusa) (Cupressaceae sensu lato)

Sugi (Cryptomeria japonica) and hinoki (Chamaecyparis obtusa) are the most important timber species in Japan. To quantify and compare the level of nucleotide variation in these species, we investigated their variation at ten nuclear loci. Average values of nucleotide diversity at synonymous sites (π _SYN) found in sugi and hinoki were 0.0038 and 0.0069, respectively. However, although the average value of nucleotide diversity was higher in hinoki than in sugi, their average values of haplotype diversity were similar. Deviations from the standard neutral model were detected at two loci in hinoki using Tajima’s D, Fay and Wu’s H, and Strobeck’s S statistics, which seem to be due to its historical population structure. Levels of divergence between the two species at synonymous sites of the ten genes ranged from 0.121 to 0.566 (0.28 on average). These values positively correlated with their guanine + cytosine contents at third-codon positions of synonymous sites (%GC3s).     

Assessment of fire blight tolerance in apple based on plant inoculations with Erwinia amylovora and DNA markers

Fire blight (Erwinia amylovora) causes serious damage to pome fruit orchards, and identification of germplasm with heritable disease resistance is therefore crucial. Two dominant SCAR (sequence characterised amplified region) marker alleles (AE10-375 and GE-8019), flanking a previously identified QTL (quantitative trait locus) for resistance to fire blight on ‘Fiesta’ linkage group 7 in apple cultivars related to ‘Cox’s Orange Pippin’, were screened on 205 apple cultivars. Both marker alleles were present in 22% of the cultivars, indicating presence of the QTL allele for tolerance, and both were lacking in 25%, indicating homozygosity for absence of the QTL tolerance allele. However, 33% had only the marker allele AE10-375, while 20% had only GE-8019, suggesting that some cultivars with the dominant alleles for both of the flanking markers can carry these on separate chromosomes and may lack the QTL allele for tolerance. In 2009 and 2010, terminal shoots of greenhouse-grown grafted trees of 21 cultivars (only 20 in 2010) were inoculated with Erwinia amylovora. ‘Idared’ (susceptible) and ‘Enterprise’ (tolerant) were included as controls. Disease severity for each cultivar was expressed as percentage of necrosis in relation to entire length of shoot, and the ranking of cultivars in 2009 and 2010 was compared with a Spearman rank correlation test, P < 0.01. A relationship between presence of both flanking marker alleles for tolerance and level of fire blight tolerance was confirmed with a Mann–Whitney U-test, P < 0.01 in 2009, and P < 0.05 in 2010. A PCO (principal coordinate) analysis based on band profiles obtained with 12 SSR (simple sequence repeat) loci produced three loose clusters, two of which contained known offspring of ‘Cox’s Orange Pippin’, and one with cultivars that were either unrelated or had an unknown origin. Cases where DNA markers did not predict level of fire blight damage as expected, were, however, as common among descendants of ‘Cox’s Orange Pippin’ as among apparently unrelated cultivars. Obviously the ‘Fiesta’ LG 7 QTL has some predictive value, both for known ‘Cox’ relatives and others, but more efficient markers would be desirable for marker-assisted selection.     

Resistance to Phytophthora fragariae var. fragariae in strawberry: the Rpf2 gene

  Phytophthora fragariae var. fragariae is the causal agent of red stele (red core) root rot in strawberry (Fragaria spp.). The inheritance of resistance to one isolate of this fungus was studied in 12 segregating populations of F.×ananassa derived from crosses between four resistant cultivars (‘Climax’, ‘Redgauntlet’, ‘Siletz’, and ‘Sparkle’) and three susceptible cultivars (‘Blakemore’, ‘Glasa’, and ‘Senga’ Sengana’). The analysis clearly supports the hypothesis of a single segregating dominant resistance gene. It is proposed that this gene be designated Rpf2. Received 12 November 1996 / Accepted: 22 November 1996     

Microsatellite Evolution: Testing the Ascertainment Bias Hypothesis

Previous studies suggest the median allele length of microsatellites is longest in the species from which the markers were derived, suggesting that an ascertainment bias was operating. We have examined whether the size distribution of microsatellite alleles between sheep and cattle is source dependent using a set of 472 microsatellites that can be amplified in both species. For those markers that were polymorphic in both species we report a significantly greater number of markers (P < 0.001) with longer median allele sizes in sheep, regardless of microsatellite origin. This finding suggests that any ascertainment bias operating during microsatellite selection is only a minor contributor to the variation observed. Received: 6 January 1997 / Accepted: 19 May 1997     

The characteristics of allelic polymorphism in killer-immunoglobulin-like receptor framework genes in African Americans

The frequencies of alleles of killer cell immunoglobulin-like receptor genes, KIR3DL3 and KIR3DL2, and the carrier frequency of KIR2DL4 alleles have been determined from a population of African Americans (n = 100) by DNA sequencing of the coding regions. Fifty alleles of KIR3DL3 were observed with the most frequent, KIR3DL3*00901 (13%). KIR3DL2 was also diverse; 32 alleles with KIR3DL2*00103 the most frequent (17%). For KIR2DL4, of the 18 alleles observed, one allele, KIR2DL4*00103, was found in 64 of the 100 individuals. Thirty-six novel alleles encoding a total of 28 unique receptors are described. Pairwise comparisons among all of the alleles at each locus suggest a predominance of synonymous substitutions. The variation at all three framework loci fits a neutral model of evolution.     

Two simple sequence repeat markers to select for soybean cyst nematode resistance coditioned by the rhg1 locus

The soybean cyst nematode (SCN) (Heterodera glycines Inchinoe) is the most economically significant soybean pest. The principal strategy to reduce or eliminate damage from this pest is the use of resistant cultivars. Identifying resistant segregants in a breeding program is a difficult and expensive process which is complicated by the oligogenic nature of the resistance and genetic variability in the pathogen. Fortunately, resistance at one SCN-resistance locus, rhg1, is generally accepted as a necessity for the development of resistant genotypes using any source of resistance and when challenged by any SCN race. Thus, the development of SCN resistant cultivars would be expedited if an effective and rapid system were available to identify breeding lines carrying a resistance allele at the rhg1 locus. In this study we report two simple sequence repeat (SSR) or microsatellite loci that cosegregate and map 0.4 cM from rhg1. Allelic variation at the first of these loci, BARC-Satt309, distinguished most, if not all, SCN-susceptible genotypes from those carrying resistance at rhg1 derived from the important SCN-resistance sources ’Peking’, PI 437654, and PI 90763. BARC-Satt309 was also effective in distinguishing SCN resistance sources PI 88788 and PI 209332 from many, but not all, susceptible genotypes. BARC-Satt309 cannot be used in marker-assisted selection in populations developed from typical southern US cultivars crossed with the important resistance sources PI 88788 or PI 209332 because these genotypes all carry the identical allele at the BARC-Satt309 locus. A second SSR locus, BARC-Sat_168, was developed from a bacterial artificial chromosome (BAC) clone that was identified using the primers to BARC-Satt309. BARC-Sat_168 distinguished PI 88788 and PI 209332 from southern US cultivars such as ’Lee’, ’Bragg’ and ’Essex’. Both BARC-Satt309 and BARC-Sat_168 were used to assay lines from SCN-susceptible×SCN-resistant crosses and proved to be highly effective in identifying lines carrying rhg1 resistance from those carrying the allele for SCN susceptibility at the rhg1 locus. Received: 5 November 1998 / Accepted: 3 February 1999     

Genetic analysis of the dwarfing gene (Rht8) in wheat. Part I. Molecular mapping of Rht8 on the short arm of chromosome 2D of bread wheat (Triticum aestivum L.)

 Two sets of single chromosome recombinant lines comparing 2D chromosomes from the wheat varieties ‘Ciano 67’ and ‘Mara’ with the common 2D chromosome of ‘Cappelle-Desprez’ in a ‘Cappelle-Desprez’ background were used to detect a diagnostic wheat microsatellite marker for the dwarfing gene Rht8. The genetic linkage maps place the wheat microsatellite marker WMS 261 0.6 cM distal to Rht8 on the short arm of chromosome 2D. By PCR analysis the WMS 261 alleles of ‘Mara’, ‘Cappelle-Desprez’ and ‘Ciano 67’ could be distinguished by different fragment sizes of 192 bp, 174 bp and 165 bp, respectively. A screen of over 100 international varieties of wheat showed that the three allelic variants were all widespread. It also demonstrated that a limited number of varieties carried novel WMS 261 variants of over 200 bp. Following classification of the individual recombinant lines for allelic variants at the WMS 261 locus it was possible to attribute a 7- to 8-cm reduction in plant height with the WMS 261-192-bp allele compared to the WMS 261-174-bp allele in the set of recombinant lines comparing 2D chromosomes of ‘Mara’ and ‘Cappelle-Desprez’. A height reduction of around 3 cm was detected between the WMS 261-174-bp allele and the WMS 261-165-bp allele in the recombinant lines comparing 2D chromosomes of ‘Cappelle-Desprez’ and ‘Ciano 67’. Received: 17 October 1997 / Accepted: 12 November 1997     

Development and characterization of 11 polymorphic compound tetranucleotide microsatellite loci for the Leisler’s bat, Nyctalus leisleri (Vespertilionidae, Chiroptera)

Eleven polymorphic microsatellite marker loci were developed from a Leisler’s bat (Nyctalus leisleri) genomic enriched library. Assessment of the usefulness of these markers for population genetics studies of Leisler’s bats was carried out by screening 100 specimens sampled from eight locations in Ireland and two in Northeastern France. Both moderately and highly polymorphic marker loci were identified. Five to 28 alleles were found to be segregating per locus with observed heterozygosities values ranging from 28.4 to 94%. Initial evaluation indicates that these microsatellites will be useful for genetic based studies aiming, for instance, at parentage and population structure of Leisler’s bats.     

Testing the rare-alleles model of quantitative variation by artificial selection

The rare-alleles model of quantitative variation posits that a common allele (the ‘wild-type’) and one or more rare alleles segregate at each locus affecting a quantitative trait; a scenario predicted by several distinct evolutionary hypotheses. Single locus arguments suggest that artificial selection should substantially increase the genetic variance (Vg) if the rare-alleles model is accurate. This paper tests the ‘ΔVg prediction’ using a large artificial selection experiment on flower size of Mimulus guttatus. Vg for flower size does evolve, increasing with selection for larger flower while decreasing in the other direction. These data are consistent with a model in which flower size variation is caused by rare, partially dominant alleles. However, this explanation becomes increasingly tenuous when considered with other data (correlated responses to selection and the effects of inbreeding). A combination of modern (marker-based mapping) and classical (biometric) techniques will likely to be required to determine the distribution of allele frequencies at loci influencing quantitative traits.     

Geographical patterns of genetic variability in introduced Australian populations of the marine toad,Bufo marinus: Sorbitol dehydrogenase,Sdh

The marine toad, Bufo marinus, was introduced to Australia from Hawaii in 1935. From 1935 to 1974, the toad population expanded exponentially to occupy 584,000 km², and now has a continuous distribution from Cape York to Tweed River on the eastern coast of the continent. Genetic analysis of the population indicates a difference in allele frequency at the sorbitol dehydrogenase locus. There are two alleles segregating at the locus (NAD-Sdh^a and NAD-Sdh^b). The NAD-Sdh^aa homozygote is common in the two southern populations, but uncommon in northern populations. The north-south difference has been established in less than 25 generations.