The affected sib method. IV. Sib trios

The classical sib pair method uses the expected and observed HLA (human leukocyte antigen) haplotype sharing distribution in sib pairs, who are affected with an HLA associated disease, to make inferences about the inheritance of the disease. In this paper we present the expected HLA haplotype sharing distributions in affected sib trios, and sib pairs, from families with three or more affected sibs. The underlying model for both distributions, as for the classical sib pair method, is that disease predisposition is determined by a single allele at an HLA-linked locus. The sib trio tests of hypotheses (additive and recessive), and disease parameter estimates (additive, recessive and intermediate), can be compared with those obtained from the classical sib pair analysis. In addition, the sib trio data allow parameter estimation for a general disease model to be made, if the data fall within the bounds of the expectation. This study forms the basis of later investigations which show that haplotype sharing of affected sib trios for two susceptibility alleles (negative complementation) model, which appears appropriate for insulin dependent diabetes mellitus (IDDM), moves outside the bound of the single susceptibility expectations outlined here, whereas haplotype sharing values for sib pairs are bound by the single susceptibility allele expectations. Available Caucasian IDDM data have been analysed. The results support genetic heterogeneity of IDDM.     

The affected sib method. II. The intermediate model

An iteration procedure is outlined which uses HLA haplotype sharing data from sib pairs in which both sibs have the disease of interest. The procedure allows estimation of the degree of dominance of the HLA linked 'disease' allele, and its frequency in the population, for intermediate models where it is assumed that individuals who do not have at least one copy of the 'disease' allele do not contract the disease. Parameter estimates from sib-pair data on multiple sclerosis, insulin dependent diabetes mellitus, haemochromatosis, coeliac disease juvenile rheumatoid arthritis, and Graves' and Hasimoto's diseases are given.     

Power of the affected-sib-pair method to detect disease susceptibility loci of small effect: An application to multiple sclerosis

The distribution of marker locus identity-by-descent scores in affected sib pairs provides a powerful tool for detecting the presence of a linked non-Mendelian disease susceptibility locus. This basic approach is here extended to include a trio of sibs. A special type of sib trio consisting of two affected and one unaffected sib is investigated. It is shown that compared to affected-sib-pairs, trios with the above configuration are less efficient in detecting the presence of a linked disease susceptibility locus. When the generalized two-allele single locus model is fitted to sib pairs affected with multiple sclerosis, an estimate of the recombination fraction of 0.21 between the putative disease susceptibility locus and the HLA complex is obtained. However, this transmission model is deemed inadequate since a recombination fraction this large is inconsistent with the variety of HLA associations observed at the population level.     

The effects of reduced fertility, method of ascertainment, and a second unlinked locus on affected sib-pair marker allele sharing

The method of affected sib-pair marker allele sharing has been used not only for the detection of linkage, but also for discerning inheritance. The application of this method has included the tacit assumption of no selective disadvantage for affected individuals. For some of the disorders analyzed by this method, this assumption is not tenable. Also, the method of ascertainment of affected sib pairs is not taken into account. It is shown here that the ascertainment procedure and reduced reproductivity of affected individuals can alter the expected distribution of affected sib-pair marker allele sharing. The effect is greatest for a recessive disease susceptibility locus. While the effect of reduced reproductivity is to inflate the gene frequency necessary to account for the observed distribution, a second unlinked disease susceptibility locus can diminish this effect. Application to type 1 diabetes shows that according to a simple recessive model with 50% penetrance, the observed distribution of affected sib-pair HLA haplotype sharing requires a gene frequency in the range of 0.55-0.60, much greater than estimates previously proposed. According to such a model, the frequency of disease would be around 15%, clearly out of range of the observed frequency of 0.4%.     

Predictions of a 2-locus model for disease heterogeneity: application to adrenoleukodystrophy.

Adrenoleukodystrophy (ALD) is an X-linked disorder that exhibits a wide range of phenotypic variability within individuals in a single family carrying the mutant allele. A 2-locus epistatic model has been proposed to explain the inheritance of the severe childhood form of ALD and the milder adult-onset adrenomyloneuropathy (AMN). Under a dominant epistatic model, a single M allele at an autosomal modifier locus ameliorates the most severe effects of the disease allele leading to the milder AMN phenotype; only males with genotype mm would have ALD. Under a recessive epistatic model, 2 copies of the M allele would be necessary to have the milder AMN phenotype. Here, we show that recurrence risks for a second affected male depend on the frequency of the protective allele at this modifier locus. Whereas it is most likely that 2 affected brothers will be concordant for their disease phenotypes, discordant pairs of affected brothers are possible at all frequencies of M. Within a narrow range of modifier allele frequencies, the predicted distribution of affected sib pairs (over all families) is consistent with empiric data from a large clinic population. Here we suggest sampling discordant affected sib pairs as a strategy for detecting linkage between a polymorphic DNA marker and a possible modifier gene. Since both epistatic models predict that discordant affected pairs should not share 2 alleles at the modifier locus, we expect that departures from the null distribution could be detected with relatively small numbers of sib pairs.     

Gene dosage and susceptibility to insulin-dependent diabetes

1. All members of 33 families in which two or more sibs have insulin-dependent diabetes mellitus (IDDM) were HLA-typed. The results strongly support the hypothesis that, closely linked to the HLA region, there is a locus ( S ) for susceptibility to IDDM. We use S ^d for alleles at this locus which confer susceptibility to disease, and S ^a for all other alleles.
2. Published prevalence (recurrence) rates for relatives of an IDDM proband, which are essential for the analysis, are reviewed critically, and a new estimate of prevalence at age 30 in the general population is calculated from published data.
3. The analysis allows for greater risk (penetrance) of the genotype S ^a S ^d (two doses of S^d) than of S ^a S ^d (one dose of S^d) and for recombination between the HLA region and S. With either our own data on 26 affected sib pairs or a much larger sample assembled by pooling with published data, the maximum-likelihood estimate of the two-dose penetrance is 71%, of the one dose penetrance 6·5%. The hypothesis of differential penetrance according to gene dosage provides a significantly better explanation than either simple dominant or simple recessive inheritance.
4. Estimates of the recombination fraction between HLA and S are very sensitive to the proportion of affected sib pairs sharing neither HLA haplotype, which varies considerably among the samples pooled. With the pooled data the recombination fraction is estimated as 3%, but in view of the strong associations with particular HLA alleles seen in population data, this should probably be considered an upper limit.     

Predictions of a 2-locus model for disease heterogeneity: Application to adrenoleukodystrophy

Adrenoleukodystrophy (ALD) is an X-linked disorder that exhibits a wide range of phenotypic variability within individuals in a single family carrying the mutant allele. A 2-locus epistatic model has been proposed to explain the inheritance of the severe childhood form of ALD and the milder adult-onset adrenomyloneuropathy (AMN). Under a dominant epistatic model, a single M allele at an autosomal modifier locus ameliorates the most severe effects of the disease allele leading to the milder AMN phenotype; only males with genotype mm would have ALD. Under a recessive epistatic model, 2 copies of the M allele would be necessary to have the milder AMN phenotype. Here, we show that recurrence risks for a second affected male depend on the frequency of the protective allele at this modifier locus. Whereas it is most likely that 2 affected brothers will be concordant for their disease phenotypes, discordant pairs of affected brothers are possible at all frequencies of M. Within a narrow range of modifier allele frequencies, the predicted distribution of affected sib pairs (over all families) is consistent with empiric data from a large clinic population. Here we suggest sampling discordant affected sib pairs as a strategy for detecting linkage between a polymorphic DNA marker and a possible modifier gene. Since both epistatic models predict that discordant affected pairs should not share 2 alleles at the modifier locus, we expect that departures from the null distribution could be detected with relatively small numbers of sib pairs. © 1992 Wiley-Liss, Inc.     

Effects of age at onset on the power of the affected sib pair and transmission/disequilibrium tests

Information on the age of a patient at disease onset, an important feature of complex diseases, is often collected in studies designed to map the disease genes. Penetrance-model-free methods, requiring no specification of penetrance functions, have been used extensively for detecting linkage and association between marker and disease loci. In this paper, we conduct an analytical study to examine the effects of incorporation of age at onset information on the power of two commonly used penetrance-model-free methods, the affected sib-pair (ASP) and transmission/disequilibrium tests (TDT). Assuming a Cox model with a major gene effect for the age at onset, we quantify analytically how age at onset affects the identity by descent (IBD) probabilities, the mean IBD values, and the expected numbers of alleles transmitted from heterozygous parents to affected children under various genetic models. We show that the power of the mean IBD test and the TDT can be greatly affected by the ages at onset of affected siblings or children used in the study. Generally, the most powerful test for ASPs is that based on affected sib pairs both having early disease onset and for TDT analyses is that based on trios with early-onset children. Naively combining affected sib pairs with different ages at onset or parent-children trios with different ages at onset of affected children can result in reduced power for detecting linkage or association. These results may be used to guide collection and analysis of sib pairs or families for diseases with variable age at onset.     

Use of HLA marker associations and HLA haplotype linkage to estimate disease risks in families with gluten-sensitive enteropathy

Based on a two-locus, double recessive model, we derive formulas for the risks that relatives of individuals with gluten-sensitive enteropathy (GSE) will also develop the disease. The calculations take advantage of: the linkage between the HLA locus and one of the two proposed GSE loci, and the preferential association of the HLA-DR3 and DR7 alleles with the GSE disease allele that occupies the HLA-linked locus. We use Bayes' rule to quantitate the strength of the association between the GSE disease allele and the HLA marker allele. This method predicts that siblings of the proband have an overall 10% risk for GSE, which is consistent with observed family data. This predicted risk rises to 30% when siblings are HLA-identical to the proband (also consistent with observed data) or when the sibling has the DR3 allele in the HLA haplotypes not shared with the proband. In those populations where DR7 also is associated with GSE, siblings of probands have a 10% predicted risk for GSE when only one HLA haplotype is shared with the proband and DR7 is included in the unshared haplotype. Other DR alleles are associated with much lower disease risks. By separating individuals into high and low risk groups, HLA typing identifies those individuals who would benefit from further diagnostic procedures. This general strategy should be applicable to other multilocus, marker-associated diseases.     

Increased sharing of maternal HLA haplotypes among children exposed to diphenylhydantoin during pregnancy.

During investigation of HLA types among children exposed to diphenylhydantoin (DPH) in utero, we found no evidence of a distortion in haplotype sharing among affected sib pairs. Unexpectedly, however, we found a marked increase in the proportion of all sib pairs (not just affected ones) sharing maternal haplotypes. Among 14 two child families, 12 shared the maternal haplotype (expected would be seven); among families with more than two children the distortion was also pronounced. This finding, if verified in future studies, could indicate that something in the mothers, whether DPH use during pregnancy, or some genetic factor associated with seizures, or some effect of the seizures themselves, may be leading to non-random segregation of HLA haplotypes in their offspring.     

HLA Haplotypes in Familial Graves'Disease

In order to further elucidate the genetics of Graves'disease, we studied two families with several affected members, as well as tested the degree to which HLA haplotypes were shared in affected sibpairs. Further, we sought to identify the disease related haplotypes by determining the haplotypes shared among affected parent-child combinations. In one family, two affected sibs differed at four possible parental HLA haplotypes; no evidence of recombination was observed which could account for the result. In the other family, five siblings were affected. Four out of the five affected sibs shared the maternal haplotype HLA-A11, Bw51, Cw5, Cw-, DRw5, Bfs, GLO1, whereas three shared the paternal haplotype HLA-Al, B8, Cw-, DRw3, Bf and GLO1. Looking at haplotype sharing, two pairs of sibs were found to be HLA identical, whereas the fifth sib shared one haplotype with one of these pairs but not with the other. Out of 14 (eight of our own and six from the literature) affected sibpairs examined, nine were found to be HLA identical and four shared one haplotype, suggesting that the contribution of both paternal haplotypes may be necessary for the susceptibility to the disease. Fourteen parent-child combinations were studied; in only three out of 13 in which the shared haplotype could be ascertained was the haplotype B8 positive; this distribution is similar to controls. However, of the remaining 10 combinations which did not share a B8 positive haplotype, five were B8 positive at one or the other of the nonshared haplotypes.     

Immunogenetic studies in families with rheumatoid arthritis and autoimmune thyroid disease.

HLA and Gm typing were carried out in 16 families. Seven families included 10 sib pairs with rheumatoid arthritis (RA) and autoimmune thyroid disease (ATD) respectively, and nine families included 16 sib pairs with RA and circulating thyroid autoantibodies respectively. Eight, 11, and seven sib pairs with either RA or clinical or immunological evidence of ATD shared none, one, and two HLA haplotypes respectively, and two, seven, and two informative sib pairs shared none, one, and two Gm haplotypes respectively. This random haplotype sharing of HLA and Gm haplotypes suggests that non-HLA, non-Gm linked genes are likely to be involved in any genetic predisposition common to RA and ATD.     

HLA and multiple sclerosis: population and families study

Association between HLA and multiple sclerosis (MS) was investigated at the population level on 100 MS patients genotyped for HLA-A, B, C, DR and Bf, Glo, and on 155 patients phenotyped for the same HLA antigens. Association between MS and DR2 was clearly confirmed, although its strength is rather weak. No other genetic marker could be related to the disease, no haplotype nor any allelic combination could be recognized as MS specific, and antigen genotype frequencies among the diseased could not ascertain the mode of inheritance, although dominance is very likely. Computer analysis between HLA, Bf, Glo and age of the patient, sex, age of onset and evolution of MS, impairment indexes, titres of anti-DNA and anti-measles antibodies in CSF did not show any interaction. Twenty sib pairs and two trios of MS were also studied; they showed no significant distortion with the random distribution of haplotypes. DR2 gene frequency, however, was significantly higher in sib pairs showing one or two haplotypes than in HLA different affected siblings. Three crossing-overs were identified which suggest where the HLA-linked MS susceptibility (MSS) gene could be located within the HLA segment, while other epistatic MSS genes or environmental factors are likely to be important.     

Affected sib-pair analysis of the contribution of HLA class I and class II loci to development of cervical cancer

Cervical cancer is a multifactorial disease and infection by oncogenic human papilloma viruses represents the main environmental risk factor. Only a subset of infections becomes persistent and develops into cancer, implying that genetic susceptibility factors are needed for malignant progression. Here, we use a population-based cohort of affected sib-pairs (ASPs) to examine the role of the human leukocyte antigen (HLA) class I and class II loci in cervical cancer susceptibility. Analysis of 278 ASPs revealed significant excess genetic sharing for all three HLA class II loci studied, DPB1, DQB1 and DRB1, with the strongest evidence for DQB1 and DRB1. No evidence of excess sharing was observed for the HLA class I HLA-B and HLA-A loci. When the material was stratified on the basis of the DQB1*0602/DRB1*1501 susceptibility haplotype, carriers showed significant sharing for all loci, whereas non-carriers showed no evidence of excess genetic sharing at any of the loci. However, for the DPB1 locus there was no difference in allele frequency between carriers and non-carriers indicating that the effect seen in DPB1 is not simply due to linkage disequilibrium. Our results show that the HLA class II represents a major genetic susceptibility locus to cervical cancer in contrary to the class I that do not appear to have a significant impact on predisposition to the disease. The strongest class II effects are coming from the DQB1 and DRB1 loci, but the DPB1 locus also contributes to the susceptibility to cervical cancer.     

Genetics of rheumatoid arthritis

The haplotype sharing distribution in affected sib pairs are used to demonstrate the linkage of a susceptibility gene for rheumatoid arthritis (RA) to the HLA region. Family and population studies suggest heterogeneity in the etiology of RA.     

An extended transmission/disequilibrium test (TDT) for multi-allele marker loci

The transmission/disequilibrium test (TDT) was recently introduced by Spielman et al. (1993) as a test for linkage and linkage disequilibrium. The test is based on the unequal probability of transmission of two different marker alleles from parents to affected offspring, when the marker locus and the hypothetical disease locus are linked and are in linkage disequilibrium. The probabilities of marker allele transmission to affected offspring conditional on parental genotype have been derived by Ott (1989) for a biallelic marker and a recessive disorder with no phenocopies. Here, we derive the transmission probabilities for a multi-allele marker locus and a generalized single locus disease model in a random sample of affected individuals from a randomly mating population. The form of these transmission probabilities suggests an extension of the TDT to multi-allele marker loci, in which the alternative hypothesis is restricted to take account of the likely pattern of unequal transmission when the recombination fraction is near 0. We show how our extended TDT can be implemented by standard software for logistic regression, although we have also written our own program which is available on request. We have evaluated the approximate power of the test under a range of realistic assumptions, and it appears that the test will often have good power when linkage disequilibrium is strong and if the disease is recessive.     

HLA haplotype segregation in families with allergic asthma

In order to evaluate genetic linkage between Allergic Asthma (AA) and the HLA system, we studied 20 families having AA affected sib pairs and 8 families with Intrinsic Asthma (IA) affected sib pairs. All AA patients had a strong IgE immune response to the mite D. farinae. Serological HLA typing (A, B, C, DR, and DQ antigens) was performed by the standard microcytotoxicity assay. Genetic analysis was made by means of the "Affected sib pairs" method. Out of the 20 affected sib pairs, 14 shared two HLA haplotypes, five shared one HLA haplotype and one was HLA-non-identical. These results differed significantly from the random ratio 1:2:1 for sharing, 2, 1 or 0 haplotypes (p less than 0.0001), and was very close to that expected for a recessive mode of inheritance. In contrast, among the IA sib pairs there was not an important distortion in the pattern of haplotypes segregation. However, a significant association between any of the HLA alleles and the two types of Asthma studied was not found. The results suggest the existence of an HLA-linked recessive gene controlling the IgE immune responsiveness to mite allergens and conferring susceptibility to AA.     

Contribution of the MHC region to the familial risk of coeliac disease.

Susceptibility to coeliac disease is genetically determined by possession of specific HLA-DQ alleles, acting in concert with one or more non-HLA linked genes. The pattern of risk seen in sibs and twins in coeliac disease is most parsimonious with a multiplicative model for the interaction between the two classes of genes. Based on a sib recurrence risk for coeliac disease of 10% and a population prevalence of 0.0033, the sib relative risk is 30. To evaluate the contribution of the MHC region to the familial risk of coeliac disease, we have examined haplotype sharing probabilities across this region in 55 coeliac disease families. Based on these probabilities the sib relative risk of coeliac disease associated with the MHC region is 3.7. Combining these results with published data on allele sharing at HLA, the estimated sib relative risk associated with the MHC region is 3.3. Therefore, the MHC genes contribute no more than 40% of the sib familial risk of coeliac disease and the non-HLA linked gene (or genes) are likely to be the stronger determinant of coeliac disease susceptibility.     

Unexpected HLA haplotype sharing in dizygotic twin pairs discordant for rheumatoid arthritis.

Dizygotic twins are generally believed to be no more genetically similar than sibs born from separate pregnancies. In the present study, a panel of 93 dizygotic twin pairs discordant for rheumatoid arthritis were typed for HLA-A, -B, -Cw, and -DR antigens. HLA haplotype sharing identical by descent between the twins showed a trend towards increased sharing of both HLA haplotypes; this increased sharing was statistically significant when the female/female twin pairs were considered separately. In contrast, the pattern of HLA haplotype sharing in sib pairs (n = 128) was consistent with a 1:2:1 ratio of 2, 1, or 0 haplotypes shared. An analysis of 16 normal dizygotic twin pairs was consistent with these results raising the possibility that dizygotic twins in general are genetically more similar at the HLA complex than sibs born from separate pregnancies.     

Fine-scale mapping using Hardy–Weinberg disequilibrium

Hardy–Weinberg disequilibrium (HWD) among affected individuals has recently been proposed for fine-scale mapping of disease susceptibility genes. We investigate the statistical properties of several available HWD measures and develop a new HWD measure J for fine-scale mapping. It is shown both theoretically and through simulations that the available HWD measures depend not only on the genetic distance between the marker locus of interest and the disease susceptibility locus, but also on the allele frequencies at the marker locus. On the contrary, the new measure is not affected by the allele frequencies at the marker locus under the following assumptions: (a) there is initial complete linkage disequilibrium between the marker and the disease loci, (b) there are no new mutations at the marker and the disease loci, and (c) the population under study is large. We develop a novel method to estimate the location of the disease susceptibility gene based on the HWD measure J . The estimator is robust to low mutation rates at the marker and the disease loci. We compare the standard error of the estimated disease gene loci using P _excess for case-control studies with the standard error using J for case-only studies under various disease models. The newly developed method is successfully applied to a data set on hereditary haemochromatosis (HH).