Molecular Biology and Evolution 19:1474-1482 (2002)
© 2002 Society for Molecular Biology and Evolution
Sequence Variation and Haplotype Structure at the Putative Flowering-Time Locus COL1 of Brassica nigra
*Department of Plant Biology, Swedish University of Agricultural Sciences, Uppsala, Sweden;
Department of Conservation Biology and Genetics, Evolutionary Biology Center, Uppsala University, Norbyvägen 18D, Sweden
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Motivated by a previous study indicating that polymorphism at an indel, Ind2, within the Brassica nigra COL1 gene is significantly associated with flowering time, we searched for evidence of selection in a sample of 41 complete sequences of B. nigra COL1. The within–gene population recombination rate is moderate, and all neutrality tests used in the present study failed to detect departure from the standard neutral model or evidence of selection. The haplotype structure of the 5'-half of the gene is primarily associated with the demographic history of the species and more specifically with the split between European and Ethiopian populations, whereas the structure of the 3'-half reflects the polymorphism at Ind2. This could be the result of selection or a combination of recombination and migration during the history of the sample of sequences. Without additional information on polymorphism in flanking areas, these two alternatives are difficult to tell apart. If selection acted on the gene, we suggest that if the indel itself is not the target of selection, among the polymorphic sites cosegregating with the polymorphism at Ind2, replacement polymorphisms around sites 890 and 1260 are the most likely quantitative trait nucleotides within the gene.
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Even if, as suggested by Gillespie (2000)
, every site in the genome is in some way affected by selection, variation at some genes is yet more likely to be altered by natural selection than at others because some genes are more closely related to individual fitness. Genes controlling flowering time certainly belong to this category: flowering too early or too late may not only expose an individual to the vagaries of the climate but, in outcrossing species, it will also affect mating success if other individuals in the population flower at a different time. Intense natural selection on a given quantitative trait, however, does not necessarily imply strong natural selection at all loci controlling that trait. First, polymorphism for a given trait in natural populations could have resulted mainly from selection at a few regulatory genes, variation at structural genes being predominantly neutral (Purruganan 2000
). Second, a large number of genes, belonging to at least three main pathways, are involved in the control of flowering time (Simpson, Gendall, and Dean 1999
) and, possibly, selection intensity at each of these genes is limited. Furthermore, most studies have generally suggested a skewed distribution of gene effects over quantitative trait loci (QTL), with a few genes with strong effects and a large number with small effects (Falconer and McKay 1996, p. 371
).
In the black mustard, Brassica nigra, a major QTL or, more correctly, a genomic region, explains up to 35% of the variation in flowering time (Lagercrantz et al. 1996
). Two candidate genes were found in this genomic region, namely, B. nigra homologs to Arabidopsis CONSTANS (Bni COa) and CONSTANS-LIKE 1 (Bni COL1). In Arabidopsis, CO acts in the pathway that accelerates flowering in response to long photoperiods (Putterill et al. 1995
, 1997
). The gene activates at least four early target genes with diverse biochemical functions that act to promote flowering, making CO a key component in the regulation of flowering time in response to the environment (Samach et al. 2000
). No variation was, however, found at Bni COa, whereas polymorphism in Bni COL1 was associated with flowering time (Kruskopf Österberg et al. 2002
). More precisely, variation at Ind2, an indel with two alleles of different lengths, was correlated to flowering time in a set of eight populations from Europe and Ethiopia. Individuals homozygous for the short allele flowered earlier than individuals carrying one or two copies of the long allele. The short allele was fixed in the southernmost populations, whereas the two alleles were segregating at higher latitudes. This observation alone does not imply a causal relationship between flowering time and the polymorphism at Ind2 because the association could as well have been caused by polymorphism at linked nucleotides. As a matter of fact, a limited study of polymorphism along the Bni COL1-COa area suggests that the intergenic space between the two adjacent genes could also be involved in the control of flowering time (Kruskopf Österberg et al. 2002
).
In the present study, we analyzed nucleotide variation in 41 complete sequences of the Bni COL1 gene. We were interested in detecting selection along Bni COL1. Tests for detecting positive or purifying selection at the intraspecific level all have limitations (Nielsen 2001
). Tests based on the allelic distribution or level of variability, such as Tajima's D-test, lack power and do not specifically test for selection but instead for a null hypothesis that also includes assumptions on the demographics of the population. Because complete selective sweeps and severe bottlenecks will lead to similar genealogies, they will also give similar values of Tajima's D, and the test will fail to tell them apart. Other genealogy-based tests such as those based on linkage disequilibrium (Kelly's ZnS test and Wall's B and Q tests) suffer from the same limitations. In contrast, tests based on comparing variability in synonymous and nonsynonymous sites, such as the McDonald and Kreitman test, are only sensitive to demography through its interaction with selection because the two types of site are interspersed in the same coding region. In principle, they require the absence of recombination, but in practice, they are only trivially affected by moderate recombination levels. Hence, we shall start by assessing some of the assumptions used by the tests, notably the absence of recombination and population structure.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
The Species
The black mustard, B. nigra (2n = 16), is an outcrossing annual, closely related to Arabidopsis thaliana. Together with other Brassica species, it likely descends from a hexaploid ancestor followed by extensive rearrangements, making its genome essentially a triplicated A. thaliana genome (Lagercrantz 1998). In temperate regions, B. nigra was a major mustard crop before the 1950s but has been replaced by B. juncea and is today mainly a weed. However, it is still used as a condiment in India and Ethiopia. Even if B. nigra is not currently the object of extensive breeding, its use as a condiment certainly implies a proportionally strong human influence. Its current natural distribution area ranges from the Atlantic Ocean eastward to India and from Southern Scandinavia southward to Ethiopia. It has also been introduced in North America. Its phylogeography and recent history are still poorly documented, but a previous study using nuclear microsatellites delineated three main groups—European, Indian, and Ethiopian (Westman and Kresovich 1999
). At a finer scale, preliminary results based on chloroplast DNA variation (P. Sjödin et al. unpublished data) also suggest admixture in populations from Northern Europe, with parental populations located in the Iberian Peninsula, Italy, and the Balkans.
Population Samples
All entries but one (Germany) were ex situ germ plasm accessions. Seed samples originating from Ethiopia (12 plants, accession number [acc. no.] BRA 1163), Greece (three plants, acc. no. BRA 185; three plants, acc. no. BRA 187), and Italy (six plants, acc. no. BRA 1164) were obtained from the Institut für Pflantzengenetik und Kulturpflanzenforschung, Gatersleben, Germany. Seed samples from Portugal (three plants, Estalagem do Gado Bravo) and Spain (two plants, Vejer de la Frontera) were kindly provided by Professor C. Gomez-Campo, and seeds from Germany (12 plants, Saale [11°52'44''E, 51°32'10''N] near Halle) were provided by Professor W. Durka. In most cases, the samples are from natural populations, although a feral origin cannot be ruled out. A single individual of A. thaliana (accession Ler) was used as an outgroup.
Molecular Methods
Genomic DNA was prepared from leaf samples as described by Liscum and Oeller (1997). For sequencing, we amplified genomic fragments (around 1,320 bp) including the single B. nigra COL1 exon and 250 bp of 5' untranslated sequence using primers CO 105 (5'-GTTGATGGGTCCTACAGAGAGAGAG-3') and CO 106 (5'-GGAGTATTATGAGAATGAAGGAACAAT-3'). Amplified products were treated with ExoSAP-IT (USB Corporation) and sequenced directly on an ABI 377 automatic sequencer (Perkin-Elmer). Polymerase chain reaction products from heterozygotes at Ind2 were cloned, and at least two clones were sequenced for each allele. The GenBank accession numbers of the sequences are AF510449–AF510489.
Analysis
Most sequence data analyses were carried out using DNAsp 5.3 (Rozas J and Rozas R 1999
). Wall's B and Q tests (Wall 1999
) and Kelly's ZnS test (Kelly 1997
) were performed and population differentiation measures calculated using PROSEQ v. 2.3 (Filatov 2001
). Both Wall's and Kelly's tests are based on the fact that selection and demography (for instance, admixture) will affect the associations among mutant alleles at different polymorphic sites. For both tests, only silent and synonymous polymorphisms were considered. Neutrality was also tested using the haplotype number test (K test) and the haplotype diversity test (H test) (Depaulis and Veuille 1998
) using the program Allelix (available from S. Mousset). As noted by Wall and Hudson (2001)
, the K test is particularly sensitive to uncertainty on the population recombination parameter, 
, and will serve to illustrate the risk associated in testing the neutral hypothesis when is assumed to be nil. The DNA Slider program (McDonald 1998
) was used to test for heterogeneity in the ratios of polymorphisms to fixed differences across COL1 using A. thaliana as an out-group.
The population recombination rate, = 4Ner, where Ne is the effective population size and r is the per locus recombination rate per generation, was estimated with the program Ldhat (McVean, Awadalla, and Fearnhead 2001
). The program implements the composite likelihood (CL) estimation method developed by Hudson (2001)
. To test whether the CL estimate is significantly different from zero, Ldhat implements a likelihood permutation test. It relies on the fact that if there is recombination, sites are not exchangeable, and consequently, the likelihood of observing the data is dependent on the order of the sites. The population recombination rate was estimated for the complete data set and for sequences from European populations only. Because the difference between the results is small, we report only the former. Finally, all between-species comparisons were carried out with the software Mega version 2.1 (Kumar et al. 2001
).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Polymorphism
The sequences were 1,320 bp long and had no introns. Two in-frame indels (Ind1 and Ind2) separated by 237 bp were found in the coding region (figs. 1 and 2 ). Ind1 is part of a trinucleotide (AAC) repeat coding for a run of asparagine residues, with five alleles (table 1 ). Remarkably, the allele with four AAC repeats was present only in the Ethiopian population, and the allele coding for four asparagine and one isoleucine residues was present in populations from both Spain and Sicily. Two alleles differing by 18 bp were detected at Ind2—the long (L) and the short (S) alleles. The allele frequencies in a more extensive survey of the polymorphism at Ind1 and Ind2 (Kruskopf Österberg et al. 2002
) are given in table 2
. The L allele was absent in the Ethiopian and Portuguese populations, but both alleles segregated elsewhere. A number of sites around Ind2 (sites 879, 886, 891, 1037, 1254, 1260, 1261, 1264, 1267, 1293) as well as one site in the noncoding part of the sequence (site 48) cosegregate with the L and S alleles, and a few sites (537, 543, 571, 583) cosegregate with Ind1.
|
|
|
|
Ignoring the cases where some nucleotides were undecided, there were a total of 47 segregating sites, of which seven were in the noncoding region (fig. 1 ). Within the coding regions, 17 changes were synonymous and 23 changes were nonsynonymous. Figure 1 gives the location of the 17 synonymous and 23 replacement sites along the gene. Although there seems to be a tendency of the replacement sites to cluster, their spatial distribution along the gene does not significantly depart from a Poisson distribution. The number of haplotypes was 26, corresponding to a haplotype diversity of 0.954 (±0.019). The nucleotide variation found in Bni COL1 is summarized in table 3 . The variation is far from uniformly distributed along the gene. Rather, there are peaks of variation close to Ind2 and at the 3' end of the gene (fig. 2 ). Finally, there was no significant codon bias (the effective number of codons is equal to 53.75).
|
Recombination and Linkage Disequilibrium
The maximum likelihood estimate of the population recombination rate,
= 4Ner, was around 3.6 (fig. 3
). The likelihood curve was rather flat for values higher than 3.6 but steep for values lower than 3.6, suggesting that this value should be considered as a minimum. A slightly smaller value was obtained when only Europe was considered. The distribution of maximum CL was obtained for 1,000 permuted data sets. No value was larger than the maximum CL for the observed data set, and we conclude that there is strong evidence of recombination (data not shown). The fit between the best fitting model and the maximum likelihood model for each pair of segregating sites was very good (data not shown). Only a few pairs showed an excess of linkage disequilibrium relative to the fitted model, and most were located around Ind2. An estimate of of the same order of magnitude, 7.1, was obtained with the Hudson (1987)
method as implemented by DNAsp.
|
Population Structure
As already observed, the polymorphism at Ind1 separates the Ethiopian populations from the European ones. Polymorphisms at some sites around Ind1 (sites 534, 537, 583) also perfectly separate Ethiopian populations from others (fig. 1 ). Considering the whole sequence also led to the conclusion that the Ethiopian population was significantly differentiated from the European ones (table 4 ). (The global fixation index KST and the KST value between the L and S allelic classes were 0.2730 and 0.1684, respectively, and both were highly significant.)
|
Testing the Standard Neutral Model
Tajima's D values were variable, mostly negative, and weakly significant (P < 0.1) at a few sites. Tajima's test was significant when it was applied to the Ethiopian sequences separately (D = -1.94, P < 0.05; see fig. 4 ) but was not significant when applied to the remaining sequences. Wall (1999)
showed that recombination rates of the order of the mutation rate, as seems to be the case in the present study, substantially reduce the power of most tests that assume the absence of recombination. Coalescent simulations assuming = 3.6 indicate that the observed mean value of Kelly's ZnS test statistic was almost significant [P(ZnS < 0.233) = 0.94]. With 26 haplotypes and a haplotype diversity of 0.954, both values were higher than the 5% significance limit of the K and H tests when was assumed to be nil and the effective population size was assumed to be 106. The latter value would correspond to a mutation rate per base pair of the order of 10-9. The same result still holds for the H test but no longer for the K test when = 3.6 (r = 6.8 x 10-9 per base pair recombination rate per generation) and Ne is kept at the same value. But if the recombination rate is increased to r = 6.8 x 10-8, none of the K and H values are significant any longer, and a number of haplotypes equal to 26 is now significantly smaller than the value expected under neutrality.
|
Divergence Between the L and S Alleles
Figure 5 gives the nucleotide diversity (
) in the S allelic class (short indel at Ind2) and nucleotide divergence (K) between the two allelic classes defined by the length of Ind2 (S vs. L). The divergence between the two alleles is most pronounced at the 3' end of the gene.
|
Between-Species Variation
Arabidopsis thaliana was used as an outgroup to test for the presence of selection through the McDonald and Kreitman (1991)
test. The silent (KS) and replacement (KA) divergences between B. nigra and A. thaliana were 0.236 and 0.105, respectively. The McDonald and Kreitman test relies on polymorphisms at silent and replacement sites and requires data from at least two species. If polymorphism within species and divergence between species are both the result of neutral mutations, the ratio of synonymous to replacement changes within species should be the same as the ratio between species. In the present case, the two ratios were not significantly different (table 5
). Finally, the ratio of polymorphism to fixed differences between B. nigra and A. thaliana when examined with McDonald's (1998)
runs test did not depart significantly from that expected under neutrality across Bni COL1. The lowest value was observed for a window centered slightly upstream of Ind2.
|
Figure 6 gives a neighbor-joining tree based on the Kimura two-parameter model. Alleles S and L form two clades, and within the S allele clade, all sequences from Ethiopia cluster together. Arabidopsis thaliana has the L allele, which is apparently absent in other Brassica species (Kruskopf Österberg et al. 2002
). This, together with the stark divergence between the two alleles, suggests that the L/S polymorphism is an ancient one that predates the separation among Brassica species.
|
The availability of one A. thaliana sequence also allows us to identify the ancestral state of polymorphic sites (fig. 1 ). In the 5'-half of the gene, the ancestral states of polymorphic sites belong either to Ethiopia or to Europe, whereas in the 3' half, ancestral sites are divided along the L/S polymorphism. The only exception is the polymorphism at site 1173 that followed, although not perfectly, the same dividing line as the polymorphism of the 5' half. Although the A. thaliana allele belongs to the L class, some S alleles are in the ancestral state at the surrounding sites. Inference of ancestral states from the A. thaliana sequence is supportive of the presence of recombination between the L and S alleles.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Motivated by a previous study indicating that polymorphism at an indel within the Bni COL1 gene was significantly associated with flowering time, we searched for evidence of selection in a sample of 41 complete sequences of Bni COL1.
Recombination
We first estimated the population recombination rate using two different methods. Wall (2000)
showed that methods to estimate have poor statistical properties, and it is therefore rather comforting to obtain estimates of the same order of magnitude. In any case, these estimates, all being single locus, coalescent-based estimates, should be taken as rough guidelines rather than at face value. A reasonable way to compare values among species is to compare the ratios /
where is the neutral mutation parameter, = 4Neµ, because in the absence of complications such as partial selfing or population structure,
|
|
/ is 0.33. The same ratio is in the range 1–8 in Drosophila (Andolfatto and Przeworski 2000
), at least 1 in humans (Przeworski and Wall 2001
), and 
1 in the Leavenworthia stylosa PgiC locus (Filatov and Charlesworth 1999
). Note that the latter belongs to the Brassicaceae family and is an outcrosser too. The ratio value is 1/10 in A. thaliana, but this low value is likely a consequence of the fact that A. thaliana is predominantly selfing, in which case a lower value is expected (Hagenblad and Nordborg 2002
). Hence, recombination within the Bni COL1 gene appears to be moderate.
Neutrality
All neutrality tests used in the present study, when applied at the whole gene level and on the complete data set, failed to detect departure from the standard neutral model (Tajima's D) or evidence of selection (McDonald and Kreitman test). This is not because of the lack of variation, because the nucleotide diversity at Bni COL1 tended to be higher than or similar to that observed in A. thaliana or closely related species (table 6 ). Brassica nigra, L. stylosa, A. lyrata, and Arabis gemmifera are outcrossers, whereas A. thaliana is predominantly selfing (99%). The power of neutrality tests is limited in the presence of recombination (Wall 1999
), but because recombination was rather low, this may not suffice to explain the lack of significant departure from neutrality. The only significant departure from the standard neutral model was obtained when the Ethiopian population was considered separately. However, the significant negative D value obtained in this case may as easily reflect a severe bottleneck as the action of selection. A bottleneck would agree with the limited genetic variation also observed in Ethiopia at microsatellite loci (Westman and Kresovich 1999
). Should we then conclude that selection did not affect the nucleotide variation at Bni COL1? Or, in other words, could the peculiar haplotype structure of Bni COL1 simply be the result of migration and recombination during the history of the sample and the absence of the L allele in Ethiopia, the consequence of strong random drift during the foundation of the Ethiopian populations? The present data are obviously not sufficient to rigorously tell apart the various scenarios that could have led to the present structure, because information at many independent loci is required to make strong inferences on past demographics and detect the presence of selection (Nordborg 2001
). Because B. nigra seeds have been and still are used in Ethiopia and India as a condiment, human-induced dispersal has likely been a major factor in the recent spread of the species, and one would perhaps have expected less structure in the distribution of the S and L alleles.
|
The L/S polymorphism
The L/S polymorphism is clearly an ancient polymorphism because the insertion was present in A. thaliana but was absent in alleles sampled from B. oleracea, B. rapa, and B. juncea (Kruskopf Österberg et al. 2002
). A deletion of the 18 bp present in A. thaliana and some B. nigra alleles could have occurred independently in the lineages leading to B. nigra, B. rapa, and B. oleracea, but a single deletion event before the split of the two lineages seems a more parsimonious explanation. This suggests that the L/S polymorphism in B. nigra predates the split of the lineage leading to the different Brassica species, an event that likely took place several million years ago. On the basis of mitochondrial nad4 genes, the divergence time for the Arabidopsis and Brassica lineages was estimated as 22 MYA (Yang et al. 1999
), and paleopalyneological data (Muller 1981
) suggest 10 MYA. In any case, recombination was frequent enough that polymorphism in the adjacent genomic area evolved independently from it (fig. 5
). Nucleotide polymorphisms cosegregating with the L/S polymorphism are found at sites 879, 886, and 891 on the 5' side of the indel, at sites 975, 1037, 1173, 1179, 1254, 1260, 1261, 1263 and 1267 on the 3' side, and in the noncoding region at site 48. Other polymorphic sites around the Ind2 region do not follow any clear pattern. Of the sites that cosegregate with the L/S polymorphism, sites 886, 891, 1037, 1254, 1260, 1261, 1263, and 1267 are replacement sites. Hence, these two groups seem the most likely candidates for being associated with flowering time. Comparison with A. thaliana COL1 sequence allows the determination of the ancestral state and indicates that sites 886 and 891 and sites 1261 and 1263 are in complete linkage disequilibrium with the L/S indel. Studies of COL1 in Arabidopsis failed to detect any effect on flowering time when COL1 was over- or underexpressed, although overexpression of COL1 did shorten the period of two circadian rhythms (Ledger et al. 2001
). Our results suggest that the situation might be different in B. nigra. CONSTANS-LIKE genes comprise two conserved domains (fig. 2
), one close to the N-terminus (coding for two B-box zinc fingers probably involved in protein-protein interaction; Robson et al. 2001
) and one close to the C-terminus which codes for a basic domain, part of which functions as a nuclear localization sequence (the CCT domain; Stryer et al. 2000
). Loss-of-function mutations in either of these domains in the CO gene showed that both domains are important for functional activity of CO gene products in A. thaliana (Robson et al. 2001
). The rates of evolution in the two conserved domains are similar to those reported for other genes, whereas other parts of the genes have evolved at an exceptionally high rate (Lagercrantz and Axelsson 2000
). All the sites cosegregating with the L/S polymorphism reside in rapidly evolving parts outside the two conserved domains. Thus, there are no functional data indicating that any of the cosegregating sites situated within the gene is a more likely candidate to be functionally associated with the variation in flowering time. Finally, a site cosegregating with the L/S polymorphism was also found in the noncoding region (site 48). In this case, as probably for sites more closely located to Ind2, physical linkage is unlikely to be the cause of the disequilibrium. The sequence pattern does not suggest any function, and additional sequencing will be needed to understand what maintains the linkage disequilibrium between this site and Ind2. The peak of variation observed at the 3'-end of the gene, together with preliminary results on the variation in the intergenic space between Bni COL1 and Bni COa, does indeed suggest that alleles in this genomic area can also be divided into two extremely divergent allele classes (O. Shavorskaya, personal communication). If our first observations are confirmed, an extensive study of linkage disequilibrium in the Bni COL1-COa genomic area will be required to understand the impact of selection on this flowering time QTL.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank the Swedish Research Council (Vetenskapsrådet grant 11166-303 to M.L.) and the Swedish Research Council for Environment, Agricultural Sciences, and Spatial Planning (FORMAS) for support. We are grateful to Alf Ceplitis, Pekka Pamilo, Outi Savolainen, Per Sjödin, and two anonymous referees for comments on the manuscript.
|
Footnotes |
|---|
Pekka Pamilo, Reviewing Editor
Keywords: Brassica nigra
COL1
flowering time
neutrality 
Address for correspondence and reprints: Martin Lascoux, Department of Conservation Biology and Genetics, Evolutionary Biology Center, Uppsala University, Norbyvägen 18D, S-752 36 Uppsala, Sweden. E-mail: martin.lascoux{at}ebc.uu.se
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Aguadé M., 2001 Nucleotide sequence variation at two genes of the phenylpropanoid pathway, the FAH1 and F3H genes, in Arabidopsis thaliana Mol. Biol. Evol 18:1-9
Andolfatto P., M. Przeworski, 2000 A genome-wide departure from the standard neutral model in natural populations of Drososphila Genetics 156:257-268
Charlesworth D., F. Liu, L. Zhang, 1998 The evolution of the alcohol dehydrogenase gene family in plants of the genus Leavenworthia (Brassicaceae): loss of introns, and an intronless gene Mol. Biol. Evol 15:552-559[Abstract]
Depaulis F., M. Veuille, 1998 Neutrality tests based on the distribution of haplotypes under an infinite-site model Mol. Biol. Evol 15:1788-1790[Web of Science][Medline]
Falconer D., T. F. C. McKay, 1996 Introduction to quantitative genetics. 4th edition Longman, London
Filatov D. A., 2001 Processor of sequences manual University of Birmingham, U.K. Available at http://www.biosciences.bham.ac.uk/labs/filatov/proseq.html
Filatov D. A., D. Charlesworth, 1999 DNA polymorphism, haplotype structure and balancing selection in the Leavenworthia PgiC locus Genetics 153:1423-1434
Gillespie J. H., 2000 Genetic drift in an infinite population: the pseudo-hitchhiking model Genetics 155:909-919
Hagenblad J., M. Nordborg, 2002 Sequence variation and haplotype structure surrounding the flowering time locus FRI in Arabidopsis thaliana Genetics 161:289-298.
Hudson R. R., 1987 Estimating the recombination rate of a finite population model without selection Genet. Res., Camb 50:245-250[Web of Science][Medline]
———. 2001 Two-locus sampling distributions and their application Genetics 159:1805-1817
Hudson R. R., D. Boos, N. L. Kaplan, 1992 A statistical test for detecting geographic subdivision Mol. Biol. Evol 9:138-151[Abstract]
Kelly J. K., 1997 A test of neutrality based on interlocus associations Genetics 146:1197-1206[Abstract]
Kruskopf Österberg M., O. Shavorskaya, M. Lascoux, U. Lagercrantz, 2002 Naturally occurring indel variation in the B. nigra COL1 gene is associated with variation in flowering time Genetics 161:299-306
Kumar S., K. Tamura, I. B. Jakobsen, M. Nei, 2001 Mega2: molecular evolutionary genetics analysis software Arizona State University, Tempe, Ariz
Lagercrantz U., 1998 Comparative mapping between Arabidopsis thaliana and Brassica nigra indicates that Brassica genomes have evolved through extensive genome replication accompanied by chromosome fusions and frequent rearrangements Genetics 150:1217-1228
Lagercrantz U., T. Axelsson, 2000 Rapid evolution of the family of CONSTANS LIKE genes in plants Mol. Biol. Evol 17:1499-1507
Lagercrantz U., J. Putterill, G. Coupland, D. Lydiate, 1996 Comparative mapping in Arabidopsis and Brassica, congruence of genes controlling flowering time Plant J 9:13-20[Web of Science][Medline]
Ledger S., C. Stryer, F. Ahton, S. A. Kay, J. Putterill, 2001 Analysis of the function of two circadian-regulated CONSTANS-LIKE genes Plant J 26:15-22[Medline]
Liscum M., P. Oeller, 1997 AFLP: not only for fingerprinting, but for positional cloning http://www-ciwdpb.stanford.edu/publications/methods/aflp.html
Liu F., L. Zhang, D. Charlesworth, 1998 Genetic diversity in Leavenworthia populations with different inbreeding levels Proc. R. Soc. Lond. Ser. B 265:293-301[Medline]
McDonald J. H., 1998 Improved tests for heterogeneity across a region of DNA sequence in the ratio of polymorphism to divergence Mol. Biol. Evol 15:377-384[Abstract]
McDonald J. H., M. Kreitman, 1991 Adaptive protein evolution at the Adh locus in Drosophila Nature 351:652-654[Medline]
McVean G., P. Awadalla, P. Fearnhead, 2001 A coalescent-based method for detecting and estimating recombination from gene sequences Genetics 160:1231-1241
Miyashita N. T., 2001 DNA variation in the 5' upstream region of the Adh locus of the wild plants Arabidopsis thaliana and Arabis gemmifera Mol. Biol. Evol 18:164-171
Miyashita N. T., A. Kawabe, H. Innan, R. Terauchi, 1998 Intra- and interspecific DNA variation and codon bias of the alcohol dehydrogenase (Adh) locus in Arabis and Arabidopsis species Mol. Biol. Evol 15:1420-1429
Muller J., 1981 Fossil pollen records of extant angiosperms Bot. Rev 47:1-142
Nielsen R., 2001 Statistical tests of selective neutrality in the age of genomics Heredity 86:641-647[Web of Science][Medline]
Nordborg M., 2001 Coalescent theory Pp.179–212 in D. J. Balding, M. Bishop, and C. Cannings, eds. Handbook of statistical genetics. John Wiley & Sons, Chichester
Przeworski M., J. D. Wall, 2001 Why is there so little intragenic linkage disequilibrium in humans? Genet. Res., Camb 77:143-151[Web of Science][Medline]
Purruganan M. D., 2000 The molecular population genetics of regulatory genes Mol. Ecol 9:1451-1461[Medline]
Putterill J., S. Ledger, K. Lee, F. Robson, G. Murphy, G. Coupland, 1997 The flowering time gene CONSTANS and homologue CONSTANS LIKE 1 (Accession no. Y10555 and Y10556) exist as a tandem repeat on chromosome 5 of Arabidopsis Plant Physiol 114:396.
Putterill J., F. Robson, K. Lee, G. Coupland, 1995 The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors Cell 80:847-857[Web of Science][Medline]
Robson F., M. M. Costa, S. R. Hepworth, I. Vizir, M. Pineiro, P. H. Reeves, J. Putterill, G. Coupland, 2001 Functional importance of conserved domains in the flowering-time gene CONSTANS demonstrated by analysis of mutant alleles and transgenic plants Plant J 28:619-631[Web of Science][Medline]
Rozas J., R. Rozas, 1999 DNAsp version 3: an integrated program for molecular population genetics and molecular evolution analysis Bioinformatics 15:174-175
Samach A., H. Onouschi, S. E. Gold, G. S. Ditta, Z. Schwarz-Sommer, M. Yanofsky, G. Coupland, 2000 Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis Science 288:1613-1616
Savolainen O., C. H. Langley, B. P. Lazzaro, H. Fréville, 2000 Contrasting patterns of nucleotide polymorphism at the alcohol dehydrogenase locus in the outcrossing Arabidopsis lyrata and the selfing Arabidopsis thaliana. Mol. Biol. Evol 17:645-655
Simpson G. G., A. R. Gendall, C. Dean, 1999 When to switch to flowering Annu. Rev. Cell Dev. Biol 15:519-550[Web of Science][Medline]
Stryer C., T. Oyama, T. F. Schultz, R. Raman, D. E. Somers, P. Más, S. Panda, J. A. Kreps, S. A. Kay, 2000 Cloning of the Arabidopsis clock gene TOC1, an autoregulatory response regulator homolog Science 289:768-771
Wall J. D., 1999 Recombination and the power of statistical tests of neutrality Genet. Res., Camb 74:65-79
———. 2000 A comparison of estimators of the population recombination rate Mol. Biol. Evol 17:156-163
Wall J. D., R. R. Hudson, 2001 Coalescent simulations and statistical tests of neutrality Mol. Biol. Evol 18:1134-1135
Watterson G. A., 1975 On the number of segregating sites in genetical models without recombination Theor. Pop. Biol 7:256-276[Web of Science][Medline]
Westman A. L., S. Kresovich, 1999 Simple sequence repeat (SSR)-based marker variation in Brassica nigra genebank accessions and weed populations Euphytica 109:85-92
Yang Y.-W., K.-N. Lai, P.-Y. Tai, W.-H. Li, 1999 Rates of nucleotide substitution in angiosperm mitochondrial DNA sequences and dates of divergence between Brassica and other angiopserm lineages J. Mol. Evol 48:597-604[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Y.-X. Yuan, J. Wu, R.-F. Sun, X.-W. Zhang, D.-H. Xu, G. Bonnema, and X.-W. Wang A naturally occurring splicing site mutation in the Brassica rapa FLC1 gene is associated with variation in flowering time J. Exp. Bot., March 1, 2009; 60(4): 1299 - 1308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lynch The Origins of Eukaryotic Gene Structure Mol. Biol. Evol., February 1, 2006; 23(2): 450 - 468. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||