Technical note: linkage disequilibrium and disease-associated CTLA4 gene polymorphisms

CTLA4 and CD28 are important regulators of T lymphocyte activation. Gene region 2q33 carrying genes for both CTLA4 and CD28 has been shown to be linked to many autoimmune diseases. Disease associations with particular CTLA4 gene polymorphisms have been reported. Recently, first lines of evidence emerged for functional effects of CTLA4 gene polymorphisms. Two independent studies reported a reduced inhibitory function of CTLA4 in individuals with certain CTLA4 genotypes: those with a high number of microsatellite repeats in one study and those with allele +49*G in exon 1 in the other one. We analyzed the strength of linkage disequilibrium between the three known CTLA4 polymorphisms among 577 independent chromosomes. Our results show that the polymorphisms previously suggested to be the functional risk factors nearly always occur together in a very frequent haplotype. Due to this strong linkage disequilibrium, we conclude that the previous reports studying merely a single polymorphism could not distinguish which variation actually caused the functional difference. Hence, either mutagenesis approaches or studies with data on all linked polymorphisms are still needed to determine the genuine functional risk polymorphism in this gene region.     

Single nucleotide polymorphisms and the linkage disequilibrium at the LDL receptor gene in Koreans

The low-density lipoprotein (LDL) receptor gene is one of the important genes in cardiovascular biology, but gives difficulty in single nucleotide polymorphism (SNP) marker choice. Twelve SNPs at the LDL receptor locus were genotyped in 117 Korean individuals. Six SNPs (rs8111982, rs12983082, rs1003723, rs5925, rs6413504, and rs2738464) were polymorphic in the Korean sample. And 8 SNPs (rs5931, rs811 1982, rs12983082, rs11669576, rs1003723, rs5925, rs641 3504, and rs2738464) were polymorphic in at least one population among the five ethnic groups (European: CEU; Chinese: CHB; Japanese: JPT; African:YRI; and Korean: KOR). Phylogenetic trees constructed based on the twelve SNPs showed a pattern clustering KOR with CHB and JPT as more closely related than either YRI or CEU. The plot of the pairwise linkage disequilibrium (LD) values for the exon 13 SNP (rs5925) against the physical distances between pairs of SNPs showed a tendency of LD decrease in the upstream of rs5925, while the low r² values are present in the downstream of rs5925 (≥0.20). The three SNPs (rs12983082, rs1003723, and rs5925) which are ethnically polymorphic are likely to be useful SNP markers for the LDL receptor locus. The interblock region between exon 13 and exon 18 delimited by the less stringent block definition (four gamete rule) contains a feasible recombination hotspot in the LDL receptor gene in Koreans.     

Single nucleotide polymorphisms and linkage disequilibrium in sunflower

Genetic diversity in modern sunflower (Helianthus annuus L.) cultivars (elite oilseed inbred lines) has been shaped by domestication and breeding bottlenecks and wild and exotic allele introgression⁻the former narrowing and the latter broadening genetic diversity. To assess single nucleotide polymorphism (SNP) frequencies, nucleotide diversity, and linkage disequilibrium (LD) in modern cultivars, alleles were resequenced from 81 genic loci distributed throughout the sunflower genome. DNA polymorphisms were abundant; 1078 SNPs (1/45.7 bp) and 178 insertions-deletions (INDELs) (1/277.0 bp) were identified in 49.4 kbp of DNA/genotype. SNPs were twofold more frequent in noncoding (1/32.1 bp) than coding (1/62.8 bp) sequences. Nucleotide diversity was only slightly lower in inbred lines (θ = 0.0094) than wild populations (θ = 0.0128). Mean haplotype diversity was 0.74. When extraploted across the genome (~3500 Mbp), sunflower was predicted to harbor at least 76.4 million common SNPs among modern cultivar alleles. LD decayed more slowly in inbred lines than wild populations (mean LD declined to 0.32 by 5.5 kbp in the former, the maximum physical distance surveyed), a difference attributed to domestication and breeding bottlenecks. SNP frequencies and LD decay are sufficient in modern sunflower cultivars for very high-density genetic mapping and high-resolution association mapping.     

Genotype determination for polymorphisms in linkage disequilibrium

Background  

Genome-wide association studies with single nucleotide polymorphisms (SNPs) show great promise to identify genetic determinants of complex human traits. In current analyses, genotype calling and imputation of missing genotypes are usually considered as two separated tasks. The genotypes of SNPs are first determined one at a time from allele signal intensities. Then the missing genotypes, i.e., no-calls caused by not perfectly separated signal clouds, are imputed based on the linkage disequilibrium (LD) between multiple SNPs. Although many statistical methods have been developed to improve either genotype calling or imputation of missing genotypes, treating the two steps independently can lead to loss of genetic information.     

Linkage disequilibrium between DNA polymorphisms within the porphobilinogen deaminase gene

Summary Three restriction fragment length polymorphisms (RFLPs) (MspI, PstI, ScrFI/BstNI) within the human porphobilinogen deaminase (PBG-D) gene have been studied in 47 unrelated patients with the autosomal dominant disorder, acute intermittent porphyria (AIP), and in 92 control subjects. Each enzyme identified a two-allele polymorphism with allele frequencies close to 0.50; however, marked linkage disequilibrium limited the number of observed haplotypes to four, of which one is uncommon. No association was detected between any haplotype and AIP.     

DNA haplotype analysis suggests linkage disequilibrium in the human insulin receptor gene

Summary Haplotypes of the insulin receptor gene were resolved in parents from Scandinavian nuclear families by studying the segregation of seven restriction fragment length polymorphisms (RFLPs). Of 97 unrelated parents, 41 had non-insulin-dependent diabetes mellitus (NIDDM). Considerable linkage disequilibrium in the region of the insulin receptor gene was found. Pairwise non-random associations were found between proximate RFLP sites, indicating the absence of recombinational hot spots between these sites. Thus, association studies between DNA polymorphisms at this locus and disease susceptibility genes could well be feasible in this population. Differences in the distribution of insulin receptor haplotypes were examined between NIDDM patients and healthy subjects. However, the differences observed were not statistically significant.     

The isolation of a genomic clone containing the apolipoprotein CII gene and the detection of linkage disequilibrium between two common DNA polymorphisms around the gene

Summary We report the isolation of a genomic clone containing the apolipoprotein CII (apo CII) gene and 5 and 3 flanking sequences. A detailed restriction map of the gene has been constructed and DNA fragments that are unique, or low copy number sequences in the genome have been identified. One of these detects a common restriction fragment length polymorphism (RFLP) with the enzyme BglI. Marked linkage disequilibrium is observed between this RFLP and that detected with the apo CII cDNA clone, even though the two variable restriction enzyme sites are approximately 12 kb apart. However, the usefulness of the apo CII gene as a marker for linkage studies is increased by the use of both RFLPs.     

Evaluation of linkage disequilibrium measures between multi-allelic markers as predictors of linkage disequilibrium between single nucleotide polymorphisms

Effectiveness of marker-assisted selection (MAS) and quantitative trait locus (QTL) mapping using population-wide linkage disequilibrium (LD) between markers and QTLs depends on the extent of LD and how it declines with distance between markers and QTLs in a population. Marker-QTL LD can be predicted from LD between markers. Our previous work evaluated LD measures between multi-allelic markers as predictors of usable LD of multi-allelic markers with QTLs. Since single nucleotide polymorphisms (SNPs) are the current marker of choice for high-density genotyping and LD-mapping of QTLs, the objective of this study was to use LD between multi-allelic markers to predict LD among biallelic SNPs or between SNPs and QTLs. Observable LD between multi-allelic markers was evaluated using nine measures. These included two pooled and standardized measures of LD between pairs of alleles at two markers based on Lewontin's LD measure, two pooled measures of squared correlations between alleles, one standardized measure using Hardy-Weinberg heterozygosities, and four measures based on the chi-square statistic for testing for association between alleles at two loci. The standardized chi-square measure that best predicted usable LD between multi-allelic markers and QTLs, based on our previous work, overestimated usable SNP-SNP or SNP-QTL LD. Instead, three other measures were found to be good predictors of usable SNP-SNP or SNP-QTL LD when LD is generated by drift. Therefore, the LD measure between multi-allelic markers that is best for predicting usable LD in a population depends on the type of markers (i.e. multi-allelic or biallelic) that will eventually be used for QTL mapping or MAS.     

Identification of tag single-nucleotide polymorphisms in regions with varying linkage disequilibrium

We compared seven different tagging single-nucleotide polymorphism (SNP) programs in 10 regions with varied amounts of linkage disequilibrium (LD) and physical distance. We used the Collaborative Studies on the Genetics of Alcoholism dataset, part of the Genetic Analysis Workshop 14. We show that in regions with moderate to strong LD these programs are relatively consistent, despite different parameters and methods. In addition, we compared the selected SNPs in a multipoint linkage analysis for one region with strong LD. As the number of selected SNPs increased, the LOD score, mean information content, and type I error also increased.     

A coalescent approach to study linkage disequilibrium between single-nucleotide polymorphisms

We present the results of extensive simulations that emulate the development and distribution of linkage disequilibrium (LD) between single-nucleotide polymorphisms (SNPs) and a gene locus that is phenotypically stratified into two classes (disease phenotype and wild-type phenotype). Our approach, based on coalescence theory, allows an explicit modeling of the demographic history of the population without conditioning on the age of the mutation, and serves as an efficient tool to carry out simulations. More specifically, we compare the influence that a constant population size or an exponentially growing population has on the amount of LD. These results indicate that attempts to locate single disease genes are most likely successful in small and constant populations. On the other hand, if we consider an exponentially growing population that started to expand from an initially constant population of reasonable size, then our simulations indicate a lower success rate. The power to detect association is enhanced if haplotypes constructed from several SNPs are used as markers. The versatility of the coalescence approach also allows the analysis of other relevant factors that influence the chances that a disease gene will be located. We show that several alleles leading to the same disease have no substantial influence on the amount of LD, as long as the differences between the disease-causing alleles are confined to the same region of the gene locus and as long as each allele occurs in an appreciable frequency. Our simulations indicate that mapping of less-frequent diseases is more likely to be successful. Moreover, we show that successful attempts to map complex diseases depend crucially on the phenotype-genotype correlations of all alleles at the disease locus. An analysis of lipoprotein lipase data indicates that our simulations capture the major features of LD occurring in biological data.     

Conserved worldwide linkage disequilibrium in the human factor XI gene

We have identified, in four diverse human populations, five common single-nucleotide polymorphisms (SNPs) in the coding region of the gene for the blood coagulation protease factor XI. Each SNP has an allele frequency >5% in at least one population. Three of the SNPs (C472T, A844G, and T1234C), spread out over approximately 10 kb of genomic DNA, are in marked linkage disequilibrium (LD) with one another (P < 10(-4)). Interestingly, haplotypes associated with the linked SNPs are conserved across all populations studied, despite significantly different allele frequencies between populations. The presence of such common, widely dispersed haplotypes could complicate the interpretation of LD studies and emphasizes the need for a better understanding of general patterns of LD to facilitate identification of genes for common disorders.     

Haplotype frequency distribution and linkage disequilibrium analysis of single nucleotide polymorphisms at the human FMO3 gene locus

We analyzed flavin-containing monooxygenase 3 (FMO3) polymorphisms, haplotype structure, and linkage disequilibrium (LD) in 256 Han Chinese and 50 African-American individuals to compare their haplotype frequencies and LD with other world populations. For the Han Chinese, genotyping of three haplotype tag single nucleotide polymorphisms (E158K, V257M, and E308G) was performed by polymerase chain reaction (PCR)-restriction fragment length polymorphism. For the African-Americans, genotyping of all coding exons was performed by modified PCR-single strand conformational polymorphism. Haplotype frequencies, LD, and evolutionary rates were inferred and estimated computationally. There were significant differences in haplotype frequency distribution and LD pattern among Han Chinese, African-Americans, and other world populations. Four major haplotypes of Han Chinese were EVE, KVE, EME, and EVG. Two major haplotypes of African-Americans were EVE and KVE. We found that sites 158 and 257 are in significant LD in both populations. This is the first report comparing FMO haplotypes and LD of Han Chinese with African-Americans. The data presented here justify further pharmacogenetic studies for potentially optimizing recommended drug dosages and evaluating relationships with disease processes.     

DMLE+: Bayesian linkage disequilibrium gene mapping

SUMMARY: The program DMLE+ allows Bayesian inference of the location of a gene carrying a mutation influencing a discrete trait (such as a disease) and/or other parameters of interest (such as mutation age) based on the observed linkage disequilibrium at multiple genetic markers. DMLE+ uses either individual marker genotypes, or haplotypes, integrates over uncertain population allele frequencies, and can incorporate prior information about gene location from an annotated human genome sequence. AVAILABILITY: DMLE+ is available in both Windows GUI and portable UNIX command line versions at http://dmle.org.     

Characterizing linkage disequilibrium and evaluating imputation power of human genomic insertion-deletion polymorphisms

Background

Indels are an important cause of human variation and central to the study of human disease. The 1000 Genomes Project Low-Coverage Pilot identified over 1.3 million indels shorter than 50 bp, of which over 890 were identified as potentially disruptive variants. Yet, despite their ubiquity, the local genomic characteristics of indels remain unexplored.

Results

Herein we describe population- and minor allele frequency-based differences in linkage disequilibrium and imputation characteristics for indels included in the 1000 Genomes Project Low-Coverage Pilot for the CEU, YRI and CHB+JPT populations. Common indels were well tagged by nearby SNPs in all studied populations, and were also tagged at a similar rate to common SNPs. Both neutral and functionally deleterious common indels were imputed with greater than 95% concordance from HapMap Phase 3 and OMNI SNP sites. Further, 38 to 56% of low frequency indels were tagged by low frequency SNPs. We were able to impute heterozygous low frequency indels with over 50% concordance. Lastly, our analysis also revealed evidence of ascertainment bias. This bias prevents us from extending the applicability of our results to highly polymorphic indels that could not be identified in the Low-Coverage Pilot.

Conclusions

Although further scope exists to improve the imputation of low frequency indels, our study demonstrates that there are already ample opportunities to retrospectively impute indels for prior genome-wide association studies and to incorporate indel imputation into future case/control studies.     

Nucleotide variability and linkage disequilibrium patterns in the porcine MUC4 gene

ABSTRACT: BACKGROUND: MUC4 is a type of membrane anchored glycoprotein and serves as the major constituent of mucus that covers epithelial surfaces of many tissues such as trachea, colon and cervix. MUC4 plays important roles in the lubrication and protection of the surface epithelium, cell proliferation and differentiation, immune response, cell adhesion and cancer development. To gain insights into the evolution of the porcine MUC4 gene, we surveyed the nucleotide variability and linkage disequilibrium (LD) within this gene in Chinese indigenous breeds and Western commercial breeds. RESULTS: A total of 53 SNPs covering the MUC4 gene were genotyped on 5 wild boars and 307 domestic pigs representing 11 Chinese breeds and 3 Western breeds. The nucleotide variability, haplotype phylogeny and LD extent of MUC4 were analyzed in these breeds. Both Chinese and Western breeds had considerable nucleotide diversity at the MUC4 locus. Western pig breeds like Duroc and Large White have comparable nucleotide diversity as many of Chinese breeds, thus artificial selection for lean pork production have not reduced the genetic variability of MUC4 in Western commercial breeds. Haplotype phylogeny analyses indicated that MUC4 had evolved divergently in Chinese and Western pigs. The dendrogram of genetic differentiation between breeds generally reflected demographic history and geographical distribution of these breeds. LD patterns were unexpectedly similar between Chinese and Western breeds, in which LD usually extended less than 20 kb. This is different from the presumed high LD extent (more than 100 kb) in Western commercial breeds. The significant positive Tajima'D, and Fu and Li's D statistics in a few Chinese and Western breeds implied that MUC4 might undergo balancing selection in domestic breeds. Nevertheless, we cautioned that the significant statistics could be upward biased by SNP ascertainment process. CONCLUSIONS: Chinese and Western breeds have similar nucleotide diversity but evolve divergently in the MUC4 region. Western breeds exhibited unusual low LD extent at the MUC4 locus, reflecting the complexity of nucleotide variability of pig genome. The finding suggests that high density (e.g. 1SNP/10 kb) markers are required to capture the underlying causal variants at such regions.     

Linkage disequilibrium and linkage analysis of the glucose-6-phosphatase gene

Recent studies have indicated that the four most common mutations account for 78% of mutant alleles in the glucose-6-phosphatase (G6Pase) gene. A significant fraction of mutant alleles remain unidentified. Thus, informative polymorphic markers are necessary for linkage analysis in carrier testing and prenatal diagnosis in families where mutations can not be identified. The common mutations appear to be ethnic-specific, suggesting that the individual mutations may have a common founder. With the recent discovery of the nucleotide 1176 polymorphism, we have studied whether these mutations are in linkage disequilibrium with the polymorphism. The results of polymerase chain reaction/allele-specific oligonucleotide analysis show that nucleotide 1176 C is in linkage disequilibrium with mutations R83 C and R83H, and with the splicing mutation 727G→T. The 1176 T polymorphism is in linkage disequilibrium with 459insTA. A GT repeat polymorphism has also been found. However, its heterozygosity is low. The 1176 nucleotide polymorphic marker can be used in carrier and prenatal diagnosis of GSD1a families that have unidentified mutations and are informative for this marker. Received: 27 January 1998 / Accepted: 17 April 1998     

Nine polymorphisms of angiotensinogen gene in the susceptibility to essential hypertension

Even if the importance of angiotensinogen (AGT) gene has been known in gene targeting animals and humans genetic studies, its precise mechanism and the interaction among AGT gene variants, plasma AGT concentration and risk for hypertension remain uncertain. We examined whether AGT gene variants predispose to hypertension via an increase of plasma AGT concentration. Plasma AGT concentration was estimated from plasma angiotensin I which was cleaved by an excess amount of human renin and measured by RIA. Using 9 AGT gene variants which included new polymorphisms (G-152A and T+31C), we examined the association with hypertension and with plasma concentration by a case-control study. Haplotype analysis revealed that G-6A, T+31C and M235T polymorphisms were in absolute linkage disequilibrium and were associated with hypertension but not with plasma AGT level. On the other hand, -1074t;T235 haplotype was associated with an increase of AGT level but not with hypertension. In the haplotype analysis, only H3 haplotype frequency, which contained G-6, T+31 and M235 alleles, was significantly increased in normotensive subjects, suggesting that this haplotype is associated with a hypotensive effect. According to combined haplotype analysis of diallele and microsatellite markers, it remains a possibility that M235T, T+31C, G-6A, A-20C and G-1074T polymorphisms may play an important role in increased risk for essential hypertension. Our results suggest that the positive association between AGT polymorphism and hypertension is not simply explained by an increase of plasma AGT concentration.