Human Molecular Genetics, 2001, Vol. 10, No. 24 2833-2839
© 2001 Oxford University Press
Long-range sequence composition mirrors linkage disequilibrium pattern in a 1.13 Mb region of human chromosome 22
Abteilung Humangenetik, Universität Ulm, Albert-Einstein-Allee 11, D-89081 Ulm, Germany
Received August 16, 2001; Revised and Accepted September 21, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Association studies, the most powerful tool for the identification of genes underlying complex traits, depend on the observation of linkage disequilibrium (LD) between marker alleles and the trait. The LD pattern of the human genome which determines the regional density of required markers is non-uniform, with regions of long-range LD over several hundred kilobases and regions where LD extends only over a few kilobases. Studying LD in the NF1 gene region we encountered a transition from long-range to short-range LD which coincides with a switch in the isochore pattern. This observation prompted us to investigate the regional variation in the extent of LD more systematically and we selected an isochore transition within the MN1/PITPNB gene region on chromosome 22q12.1. Long-range LD characterizes the GC-poor (40% GC) parts of the sequences. No LD can be observed between closely spaced markers throughout the whole range of the GC-rich (50% GC) parts. In both cases, the NF1 and the MN1/PITPNB gene region, a clear-cut transition of the long-range GC content precisely coincides with a change in the extent of observable LD. The results can be explained by a 72-fold lower recombination frequency in the GC-poor, compared to the GC-rich isochores. Although recombination is not the only factor governing LD, our findings can help to predict levels of LD and marker densities required for association studies on the basis of regional GC content.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
There has been much debate in the literature during the last years on the extent and pattern of linkage disequilibrium (LD) in the human genome. The identification of genes responsible for common human diseases, one of the major goals of human genetics in the near future, depends on LD between markers and traits. Thus, the number of markers which have to be applied in association studies depends on the distance over which LD can be detected. Starting from theoretical simulations (1,2), which resulted in a required density of one marker per 10 kb in the general human population, a number of larger studies (3–6) on the extent of LD in various genomic regions and various populations showed that the extent of LD in the human genome is non-uniform, with regions of long-range LD over several hundred kilobases and regions with no allelic association between markers at a distance of only a few kilobases. For that reason it is not useful to give an average value for the spacing of markers for association studies but an LD map has to be established for each genomic region under investigation. Knowledge of the rules guiding the local extent of LD in different regions would simplify this task considerably. One of the factors which influence LD is the recombination frequency. Cytogenetic analyses showed that chiasmata, the visible correlate of meiotic crossover, are seen more frequently in R and T bands than in G bands (7). This observation relates the recombination frequency to compositional patterns of the human genome, because R and T bands are composed primarily of GC-rich DNA sequences, classified as H2 and H3 isochores according to Bernardi (8), whereas G bands contain almost exclusively GC-poor sequences (9–11), called L1, L2 and H1 isochores. We previously showed a region with long-range LD over 350 kb at the NF1-gene locus directly neighboured by a region without LD between closely spaced markers (36 and 27 kb). The boundary between the two regions precisely coincides with a sharp transition of the long-range GC content of the underlying sequences (12). The region with long-range LD shows a GC content of 39%, the region with no LD of 51%, characteristic values for L1 and H2 isochores. This was the first hint on a possible correlation between isochore structure and LD pattern on a molecular level. However, it was a single observation and in the H2 isochore only a few markers were analysed. To see whether this observation can be generalized, we now established the LD pattern of a 1.13 Mb region from the MN1/PITPNB gene locus on chromosome 22q12.1 covering a transition from an L1 to a H2 isochore and a short interspersed fragment of high GC content within the L1 isochore (13).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
To obtain a detailed isochore map of the MN1/PITPNB gene region 10 consecutive sequence stretches from the published sequence of chromosome 22 (13) were analysed by moving a 5 kb window in steps of 5 kb over the whole sequence range. The accession numbers of the sequences are given in Table 1. The resulting GC pattern, depicted in Figure 1, shows a clear cut transition between two isochores. The centromeric part of the analysed region, 470 kb in length, has a GC content of 49–50% which is a typical score of H2 isochores. The H2 isochore is followed by a very narrow transition region of only 5 kb where the GC value drops to 39%. The telomeric part of 660 kb has a GC content of 38–40%, a typical value for L1 isochores, with the exception of a sharp 50 kb peak of 53% GC located 130 kb telomeric to the main isochore transition.
|
|
To cover the region with polymorphisms at a density of approximately 1 SNP per 30 kb we searched for appropriate sequence variants in the chromosome 22 SNP database (Sanger Centre). The accession IDs of the SNPs used for this work, the accession numbers of the respective sequences and the allele frequencies found in our population are given in Table 1.
The extent of LD was determined by establishing the length of intervals in which the |D'| value was higher than 0.5, as described in Materials and Methods. The length of the intervals, the |D'| values found between the SNP pairs and the 
2 values are given in Table 2. Regarding the extent of allelic association, the analysed region can be divided into two parts. Within a proximal part of 460 kb 15 markers define 14 intervals (1–14), 2–66 kb in length, where no or only weak allelic association can be found. Only two pairs of closely spaced markers (intervals 12 and 1 with 8 and 33 kb, respectively) show |D'| values higher than 0.3. Even between two markers only 2 kb away from each other (interval 14) no LD can be found (|D'| = 0.065). In the distal part of the region, 670 kb in length and directly adjoined to the first one, four additional markers (intervals 15–18) are located 80–295 kb away from each other. In all cases high |D'| values (from 0.616 to 0.986) between neighbouring markers can be observed despite the long distances, defining a region of long-range LD. The boundary between the sequences without LD and long-range LD is sharp. SNP stdJ353E16_68600, located at position 68600 in sequence al031591 marks the boundary (position 458 kb in Figs 1 and 2). No LD can be found with neighbouring markers located centromerically, whereas long-range LD (|D'| = 0.854) can be observed between this marker and the next marker, located 295 kb away in telomeric direction. The transition of the long-range GC content from 50 to 40% also takes place in sequence al031591 10 kb telomeric to stdJ353E16_68600 between position 75000 and 80000 (around position 468 kb in Figs 1 and 2). Therefore, the isochore boundary found in the MN1/PITPNB gene region precisely coincides with a boundary between a region with long-range LD and a region without LD. Long-range LD can be found in the GC-poor isochore (L1) and ends within 10 kb after the sharp transition to a GC-rich isochore (H2).
|
|
D' is a measure for allelic association between marker pairs or between markers and traits irrespective of the distances between the analysed loci. It cannot be normalized for physical distance, a fact which makes direct comparisons of the LD patterns of different genomic regions difficult. However, D' is directly correlated to the recombination frequency
which can be normalized for the physical distance between loci. Therefore, we calculated recombination frequencies from the |D'| of all marker pairs assuming different ages of the sequence variants (n = 1000, 3000 and 5000 generations). After correction for the physical distance (Mb) between the marker pairs, we obtained the results shown in Table 2. The /Mb values calculated for an age of the population of 3000 generations are shown graphically in Figure 2. The values for 3000 generations were chosen for the presentation because an analysis, published by Reich et al. (6), showed that LD in European populations is governed by a demographic bottleneck, which occurred 3000 generations ago or even more recently. The /Mb values found between marker pairs within the GC-rich isochore are higher than the genomic average of 1 cM/Mb and are relatively uniform, with the exception of interval 14, which is located at the boundary between the regions with long- and short-range LD. The score for this 2 kb interval is 15 times higher than the average for the GC-rich isochore, possibly indicating a localized hotspot of recombination as described for some other genomic loci (14,15). Within the GC-poor isochore the recombination frequencies per megabase are also relatively uniform but are considerably lower than the scores for the GC-rich isochore. The /Mb values given in Table 2 and shown in Figure 2 cannot be regarded as absolute estimates, due to two problems. One problem arises from statistical uncertainties in the estimation of |D'| for individual intervals, which is the basis for the calculation of /Mb. For this reason the differences in /Mb found within intervals 1–13 on the one hand and within intervals 15–18 on the other hand cannot be regarded as real differences between the recombination frequencies of individual intervals. Instead, they can be regarded as independent and repeated measurements of the recombination frequencies typical for the two larger regions, for which mean values can be calculated. For the /Mb values of intervals 1–13 a mean of 0.026 (SD = 0.013, SE = 0.004) was obtained, for intervals 15–18 a mean of 0.00036 (SD = 0.00042, SE = 0.00021). The scores obtained for the regions of long- and short-range LD differ significantly by a factor of 72 from each other (P = 0.0017). The /Mb in interval 14 is above 2 SD for single measurements. A second problem arises from the unknown history of the population and can lead to an error in the absolute score of the recombination frequencies. However, this error affects all estimates equally and, therefore, does not influence the relation of the values for the GC-rich and GC-poor isochores.
Regional differences in recombination rates of an extent found between the GC-rich and GC-poor sequence parts at the MN1/PITPNB gene locus should be visible in genetic maps. Therefore, we looked for the genetic distances between appropriate markers in the Marshfield map (16) and correlated them with the physical distances between the markers. Both maps are shown in Figure 3. Marker D22S1163 is located relatively close to the isochore transition. The next markers in the centromeric direction for which the genetic and physical map positions are consistent are D22S1144 and D22S1167, with physical distances of 0.236 and 0.89 Mb and genetic distances of 0.54 and 3.28 cM to D22S1163. In the telomeric direction two markers, D22S689 and D22S1150, are located at a genetic distance of 0.55 cM from D22S1163. The physical distances are 0.9 and 1.55 Mb, respectively. Normalized for a distance of 1 Mb these values result in recombination frequencies of 2.3 and 3.7 cM/Mb, respectively, in the GC-rich region, and 0.61 and 0.35 cM/Mb, respectively, in the GC-poor region. Thus, the map-derived values show the same tendency as the LD-derived values, but the differences between the two regions in the map are less pronounced. The map-derived recombination frequency within the GC-rich isochore is in the range of the LD-derived frequencies calculated for 3000 generations (2.6 ± 1.3 cM/Mb). For the GC-poor isochores the map-derived value is higher than the LD-derived value (0.036 ± 0.042 cM/Mb), possibly a result of the relatively low resolution of the genetic map. Marker D22S1163 is located in sequence Al050402 in the H2 isochore 300 kb away from the isochore boundary. Therefore, as can be seen in Figure 3, the regions between markers D22S1163 and D22S689 and between markers D22S1163 and D22S1150 are composed of two parts each, a common 0.3 Mb part with a high GC content and short-range LD and a 0.6 or 1.25 Mb part with a low GC content and long-range LD. For this reason the recombination frequency for the L1 isochore cannot be given independently from the 0.3 Mb H2 part.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Our results on allelic association between markers in the MN1/PITPNB gene region on chromosome 22q12.1 demonstrate that an isochore transition precisely coincides with a switch in the extent of LD. This seems to be a more general feature of the human genome, because the same observation was made in the NF1 gene region on chromosome 17q11.2 (12). In both cases long-range LD can be found in an L1 isochore and no LD in a H2 isochore. In addition, the analysis of the extended region on chromosome 22 shows that there is not only a sharp boundary between regions with and without LD, but the extent of LD is relatively uniform throughout an isochore. Small regions with a deviant GC content from the main isochore are not depicted on the LD map, as can be seen for the 50 kb stretch with high GC content in the 850 kb L1 isochore in the MN1/PITPNB gene region. The 50 kb peak is located within interval 15 where LD can be observed over 295 kb, and therefore, obviously does not lead to a breakdown of long-range LD.
The extent of LD is the result of a complex interplay between a number of factors. The history of a population, especially bottlenecks in population size and admixture of different populations, have an influence on the extent of LD (3,6,17). These factors are supposed to act on a genome-wide level and should not have a strong regional impact. They cannot explain differences of allelic association in neighbouring isochores found in the same population. Selection can create and maintain long-range LD as discussed for the MHC locus (18,19), but cannot explain sharp boundaries between regions with and without LD. A factor that directly influences the extent of LD is the recombination frequency. Recombination is highly non-uniform in the human genome as can be demonstrated on a large scale by comparing the physical and genetic maps (20,21) and on a local scale by the presence of recombination hot spots, which comprise only a few kilobases of DNA (14,15). Therefore, regional differences in the extent of LD as described for the NF1 and the MN1/PITPNB gene regions, most probably result from regional differences in the recombination frequency. In the MN1/PITPNB gene locus such differences are present in the genetic map. For the NF1 gene locus reliable values could not be given, because the genetic and physical maps are not consistent.
Our observations correlate the recombination frequency to the long-range GC pattern of the human genome. Isochores are known to be the molecular counterpart of the chromosomal structure and define the character of the underlying sequences. GC-poor and GC-rich isochores differ from each other in terms of gene density, content of repetitive elements and replication timing (7,8,20,22). The correlation between recombination frequencies and replication timing with compositional genomic patterns may not simply be the result of structure, but may have functional aspects, as both are discussed as forces responsible for the evolution and maintenance of isochores (23). Replication timing is thought to influence the mutational input and thereby, on a long range, the composition of DNA sequences (24). During the process of recombination, heteroduplex DNA is formed and mismatches within the heteroduplex may be repaired, which leads to gene conversion. This process seems to be biased and preferably leads to G or C as a result of the repair (25). Regional differences in recombination frequencies should lead to differences in the frequencies with which biased gene conversion occurs, and thereby have an influence on the regional GC content. Sharp boundaries between genomic regions with different recombination frequencies, as found in the NF1 and MN1/PITPNB gene regions, may therefore be responsible for the maintenance of sharp boundaries between isochores.
A more practical implication of our work concerns the design of association studies. The observation that the recombination frequency and hence the extent of LD is dependent on the class of an isochore raises the possibility that the marker density can be adjusted to the GC content of the underlying sequences on a large scale. However, as recombination is not the only factor influencing LD, genomic regions exist where no correlation between isochore structure and LD can be found. For example, in Xq28 a 350 kb region with long-range LD was detected in GC-rich sequences (4) and at the MHC locus allelic association between markers can be found, although a hotspot of recombination is located between them (14). Even though there are exceptions generated by factors independent from isochores, like selection, the adjustment of marker densities to the GC content should be feasible in genomic regions where LD is primarily guided by the recombination frequency and should considerably reduce the typing effort required to establish LD maps. In GC-poor regions one marker every 100 kb may be sufficient, whereas in GC-rich regions a density of one marker per 10 kb may be necessary, as suggested by theoretical simulations (1).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
SNP typing
Possible SNPs were searched in the chromosome 22 SNP database at the Sanger Centre. The accession numbers of the sequences where the SNPs are derived from and the IDs of the SNPs used in this study are given in Table 1. Sequence variations were chosen, which affected the recognition sequences of restriction enzymes, or PCR primers were designed in a way, that a restriction site was created comprising the SNP. Typing was performed by PCR amplification of the respective fragments from genomic DNA from white blood cells [isolated according to Miller et al. (26)], restriction with the appropriate enzyme and conventional agarose gel electrophoresis. To analyse whether a given variant was polymorphic the DNA of six probands was typed. Initially 130 sequence variants were tested to obtain a sufficiently dense coverage of the region with polymorphic markers. All polymorphisms used for further analysis were in Hardy–Weinberg equilibrium. When possible, polymorphisms with equal allele frequencies were chosen for further analysis, which was done with DNA from 95 unrelated individuals, selected from the southern German population. If a polymorphism had to be used with a frequency of the rarer allele lower than 0.25, this SNP and the neighbouring SNPs were typed in a second group of 95 individuals in addition.
Statistical methods
Haplotype frequencies were calculated with the EH program of Terwilliger and Ott (27), available at the Human Genome Mapping Project website. EH was also used to test the significance of association with the 2 test. D' was calculated as described (28). A LD map of the region was established by calculating D' between neighbouring SNPs starting with the most centromeric and telomeric markers as anchor markers. To get a survey of the LD pattern the following strategy was used. In the case of a |D'| score higher than 0.5 the next neighbouring marker was analysed with the anchor marker. If |D'| between a marker and the anchor marker was lower than 0.5 this marker was used as a new anchor marker. In this way the 1.13 Mb region was divided into 18 intervals with maximal extent of LD. The recombination frequency was deduced from |D'| with the formula = 1 – |D'|1/n, which is a transformation of the equation |D'| = (1 – )n (5), where n is the assumed age of the variant in generations. The values were normalized to the physical distance between markers and presented as /Mb.
|
ACKNOWLEDGEMENTS |
|---|
We would like to thank Britta Pietsch for excellent technical assistance.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed. Tel: +49 731 5002 3426; Fax: +49 731 5002 3438; Email: guenter.assum@medizin.uni-ulm.de
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Kruglyak, L. (1999) Prospects for whole-genome linkage disequilibrium mapping of common disease genes. Nat. Genet., 22, 139–144.[Web of Science][Medline]
2 Zöllner, S. and von Haeseler, A. (2000) A coalescent approach to study linkage disequilibrium between single-nucleotide polymorphisms. Am. J. Hum. Genet., 66, 615–628.[Web of Science][Medline]
3 Dunning, A.M., Durocher, F., Healey, C.S., Dawn Teare, M., McBride, S.E., Carlomagno, F., Xu, C., Dawson, E., Rhodes, S., Ueda, S. et al. (2000) The extent of linkage disequilibrium in four populations with distinct demographic histories. Am. J. Hum. Genet., 67, 1544–1554.[Web of Science][Medline]
4 Taillon-Miller, P., Bauer-Sardina, I., Saccone, N.L., Putzel, J., Laitinen, T., Cao, A., Kere, J., Pilia, G., Rice, J.P. and Kwok, P.-Y. (2000) Juxtaposed regions of extensive and minimal linkage disequilibrium in human Xq25 and Xq28. Nat. Genet., 25, 324–328.[Web of Science][Medline]
5 Abecasis, G.R., Noguchi, E., Heinzmann, A., Traherne, J.A., Bhattacharyya, S., Leaves, N.I., Anderson, G.G., Zhang, Y., Lench, N.J., Carey, A. et al. (2001) Extent and distribution of linkage disequilibrium in three genomic regions. Am. J. Hum. Genet., 68, 191–197.[Web of Science][Medline]
6 Reich, D.E., Cargill, M., Bolk, S., Ireland, J., Sabeti, P.C. Richter, D.J., Lavery, T., Kouyoumijan, R., Farhadian, S.F., Ward, R. and Lander, E.S. (2001) Linkage disequilibrium in the human genome. Nature, 411, 199–204.[Medline]
7 Holmquist, G.P. (1992) Chromosome bands, their chromatin flavors, and their functional features. Am. J. Hum. Genet., 51, 17–37.[Web of Science][Medline]
8 Bernardi, G. (2000) Isochores and the evolutionary genomics of vertebrates. Gene, 241, 3–17.[Web of Science][Medline]
9 Saccone, S., De Sario, A., Wiegant, J., Raap, A.K., Della Valle, G. and Bernardi, G. (1993) Correlations between isochores and chromosomal bands in the human genome. Proc. Natl Acad. Sci. USA, 90, 11929–11933.
10 Saccone, S., Caccio, S., Kusuda, J., Andreozzi, L. and Bernardi, G. (1996) Identification of the gene-richest bands in human chromosomes. Gene, 174, 85–94.[Web of Science][Medline]
11 Saccone, S., Federico, C., Solovei, I., Croquette, M-F., Della Valle, G. and Bernardi, G. (1999) Identification of the gene-richest bands in human prometaphase chromosomes. Chromosome Res., 7, 379–386.[Web of Science][Medline]
12 Eisenbarth, I., Vogel, G., Krone, W., Vogel, W. and Assum, G. (2000) An isochore transition in the NF1 gene region coincides precisely with a switch in the extent of linkage disequilibrium. Am. J. Hum. Genet., 67, 873–880.[Web of Science][Medline]
13 Dunham, I., Hunt, A.R., Collins, J.E., Bruskiewich, R., Bearne, D.M., Clamp, M., Smink, L.J., Ainscough, R., Almeida, J.P., Babbage, A. et al. (1999) The DNA sequence of human chromosome 22. Nature, 402, 489–495.[Medline]
14 Jeffreys, A.J., Ritchie, A. and Neumann, R. (2000) High resolution analysis of haplotype diversity and meiotic crossover in the human TAP2 recombination hotspot. Hum. Mol. Genet., 9, 725–733.
15 Badge, R.M., Yardley, J., Jeffreys, A.J. and Armour, J.A. (2000) Crossover breakpoint mapping identifies a subtelomeric hotspot for male meiotic recombination. Hum. Mol. Genet., 9, 1239–1244.
16 Broman, K.W., Murray, J.C., Sheffield, V.C., White, R.L. and Weber, J.L. (1998) Comprehensive human genetic maps: individual and sex-specific variation in recombination. Am. J. Hum. Genet., 63, 861–869.[Web of Science][Medline]
17 Wilson, J.F. and Goldstein, D. (2000) Consistent long-range linkage disequilibrium generated by admixture in a Bantu-Semitic hybrid population. Am. J. Hum. Genet., 67, 926–935.[Web of Science][Medline]
18 Carrington, M. (1999) Recombination within the human MHC. Immunol. Rev., 167, 245–256.[Web of Science][Medline]
19 Sanchez-Mazas, A., Djoulah, S., Busson, M., Le Monnier de Gouville, I., Poirier, J.C., Dehay, C., Charron, D., Excoffier, L., Schneider, S., Langaney, A., Dausset, J. and Hors, J. (2000) A linkage disequilibrium map of the MHC region based on the analysis of 14 loci haplotypes in 50 French families. Eur. J. Hum. Genet., 8, 33–41.[Web of Science][Medline]
20 International Human Genome Sequencing Consortium (2001) Initial sequencing and analysis of the human genome. Nature, 409, 860–921.[Medline]
21 Yu, A., Zhao, C., Fan, Y., Jang, W., Mungall, A.J., Deloukas, P., Olsen, A., Doggett, N.A., Ghebranious, N., Broman, K.W. and Weber, J.L. (2001) Comparison of human genetic and sequence-based physical maps. Nature, 409, 951–953[Medline]
22 Tenzen, T., Yamagata, T., Fukagawa, T., Sugaya, K., Ando, A., Inoko, H., Gojobori, T., Fujiyama, A., Okumura, K. and Ikemura, T. (1997) Precise switching of DNA replication timing in the GC content transition area in the human major histocompatibility complex. Mol. Cell. Biol., 17, 4043–4050.[Abstract]
23 Eyre-Walker, A. and Hurst, L.D. (2001) The evolution of isochores. Nat. Rev. Genet., 2, 549–555.
24 Wolfe, K.H., Sharp, P.M. and Li, W.-H. (1989) Mutation rates differ among regions of the mammalian genome. Nature, 337, 283–285.[Medline]
25 Brown, T.C. and Jiricny, J. (1988) Different base/base mispairs are corrected with different efficiencies and specificities in monkey kidney cells. Cell, 54, 705–711.[Web of Science][Medline]
26 Miller, S.A., Dykes, D.D. and Polesky, H.F. (1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res., 16, 1215.
27 Terwilliger, J.D. and Ott, J. (1994) Handbook for Human Genetic Linkage. John Hopkins University Press, Baltimore, PA.
28 Devlin, B. and Risch, N, (1995) A comparison of linkage disequilibrium measures for fine-scale mapping. Genomics, 29, 311–322.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Schmegner, J. Hoegel, W. Vogel, and G. Assum The Rate, Not the Spectrum, of Base Pair Substitutions Changes at a GC-Content Transition in the Human NF1 Gene Region: Implications for the Evolution of the Mammalian Genome Structure Genetics, January 1, 2007; 175(1): 421 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Vinogradov DNA helix: the importance of being GC-rich Nucleic Acids Res., April 1, 2003; 31(7): 1838 - 1844. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. D. Petes and J. D. Merker Context Dependence of Meiotic Recombination Hotspots in Yeast: The Relationship Between Recombination Activity of a Reporter Construct and Base Composition Genetics, December 1, 2002; 162(4): 2049 - 2052. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Heutink and B. A. Oostra Gene finding in genetically isolated populations Hum. Mol. Genet., October 1, 2002; 11(20): 2507 - 2515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Birdsell Integrating Genomics, Bioinformatics, and Classical Genetics to Study the Effects of Recombination on Genome Evolution Mol. Biol. Evol., July 1, 2002; 19(7): 1181 - 1197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Tiret, O. Poirier, V. Nicaud, S. Barbaux, S.-M. Herrmann, C. Perret, S. Raoux, C. Francomme, G. Lebard, D. Tregouet, et al. Heterogeneity of linkage disequilibrium in human genes has implications for association studies of common diseases Hum. Mol. Genet., February 1, 2002; 11(4): 419 - 429. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||