The Recombination Fraction of the HL-A System

By collecting all the family material available in the seven Danish, Norwegian, and Swedish centres typing for Scandiatransplant, we found 309 families which yielded information on linkage between the first and second segregant series of the HL-A transplantation system. Information was obtained on 1,362 parental meiotic divisions and recombination has occurred in 11 cases which gives a recombination fraction of 0.0081 with 95% confidence limits of 0.0041–0.0148. The maternal recombination frequency (0.0099) was higher than the paternal one (0.0061), but the difference did not reach statistical significance. Most of the recombinant children belonged to the last half of their sibship.     

The Gm-Pi linkage heterogeneity in view of Pi M subtypes

In this study linkage between the loci for Gm (γ-type heavy-chain immunoglobulin markers) and Pi (α₁-antitrypsin/α₁-protease inhibitor) has been shown in families segregating for the Pi M subtypes (Ml, M2, M3 and Msal) as identified by separator isoelectric focusing. The estimate for the Gm-Pi (M-type) recombination is 0-29 (95% limits 0-24-O37) at a peak lod score of 4-31 and with no sex difference. This value is not significantly different from updated recombination frequency estimates for Gm-Pi in Pi MS (0-26) and Pi MZ, SZ and FZ families (0 21). The overall Gm-Pi recombination fraction estimate of 0 26 (95 % limits O23-0-30) at a peak lod score of 20-75 must now be considered as solid. There is a significant heterogeneity within the male Pi MZ families in that the new Finnish families show a higher recombination between Gm and Pi. There is also a possible segregation distortion (Z:M = 23:8). The heterogeneity is discussed in terms of haplotypes, the behaviour of which could be determined by linked genes or chromosomal rearrangements. The possibility that the α₁-antitrypsin level influences recombination frequency has not been ruled out, but cannot explain the heterogeneity within Pi MZ families.     

An estimation of the recombination fraction between the MLC locus and the FOUR locus.

Thirty-nine families with 167 children were typed for HL-A and studied in the MLC test. Two maternal recombinations between the FOUR and the MLC locus in two different families were found. An estimate of the recombination fraction between the FOUR and the MLC locus was calculated.     

Utilizing automated methods to improve estimates of recurrence risk with linked genetic markers

Information on linked markers can greatly improve estimates of recurrence risk for genetic disorders. However, when linkage is incomplete, utilizing the information in genetic counseling requires complex, repetitive calculations to estimate correct risks whenever the parental linkage phase is not precisely known. This paper demonstrates with examples how the computer program LIPED can be used for any pedigree to estimate risk precisely within bounds determined only by confidence limits on the recombination fraction.     

An Estimation of the Recombination Fraction between the MLC Locus and the FOUR Locus

Thirty-nine families with 167 children were typed for HL-A and studied in the MLC test. Two maternal recombinations between the FOUR and the MLC locus in two different families were found. An estimate of the recombination fraction between the FOUR and the MLC locus was calculated.     

Linkage between the loci for benign (Becker-type) X-borne muscular dystrophy and deutan colour blindness

A family is described in which benign Becker type X-linked muscular dystrophy and deutan colour blindness are segregating. The lod scores from this family have been added to those obtained in a family previously reported (Emery et.al, 1968/1969) and give an estimate of 0·23 for the recombination fraction with 95% confidence limits of 0·13 to 0·43. These results confirm the linkage relationships between deutan colour blindness and Becker muscular dystrophy but since the loci for Duchenne muscular dystrophy and colour blindness are not within measurable distance of each other these results indicate that the Becker and Duchenne types of X-linked muscular dystrophy are not allelic.     

New linkage data for the X-linked types of muscular dystrophy and G6PD variants, colour blindness, and Xg blood groups

Six families in which Duchenne muscular dystrophy (DMD) and G6PD or deutan colour blindness are segregating are reported. The sum of the lod-scores of these families together with three published previously indicates that the DMD locus is far from the G6PD:deutan cluster. The lod-scores of two families with Becker muscular dystrophy (BMD) informative for the G6PD locus together with those of one family previously studied by Emery, Smith, and Sanger (1969) suggest that the BMD locus could be at a measurable distance from this cluster. The maximum likelihood estimate of the recombination fraction is 0·27 and the 90% confidence limits are 0·17 and 0·40. This difference in linkage estimates for DMD and BMD suggests that the BMD and the DMD genes are located at two different loci on the X chromosome.

Five more families with DMD and two with BMD informative for Xg blood groups support the conclusion of other authors that there is no hint of linkage between the loci for Xg and for the X-linked forms of muscular dystrophy.

     

A Study on HL—A System in Japanese

Two hundred unrelated Japanese individuals were HL-A typed with UCLA Research Tray T3 (Terasaki's Tray), which contains specificities added after the Fifth International Workshop. Phenotype, gene and haplotype frequencies were calculated with standard errors and delta values. HL-A9, HL-A5 and W10 had a higher frequency and HL-A1, 3 and 8 had a lower frequency in Japanese than in Caucasians. The frequent haplotypes were HL-A9-HL-A5, HL-A9-HL-A7 and HL-A2W10. HL-A9-HL-A5 showed very positive high linkage disequilibrium parameter (delta value) and HL-A9-W10 showed negative high value. The sera designated as anti-HL-A, W5 and W15 in the T3 Tray which react identically in Caucasians showed different patterns of reaction when tested in the Japanese population. Five hundred Japanese parous women's sera were tested for cytotoxic antibodies. Some Japanese antisera showed high correlation coefficient values on HL-A2, HL-A9, HL-A10, HL-A11 and HL-A12. The women providing the anti-HL-5 complex sera and their immunizing persons were HL-A typed. These complex sera reactions were compared with the antisera in the T3 Tray. A new group of sera (SN-1), "operationally monospecific" and cross-reacting with W22, was found in the present study. Population and family studies suggested that the sera SN-1 are third in frequency within the second series (phenotypic frequency 17-22%) and show high delta values with HL-A11 in the Japanese population.     

Very close linkage between D2S1 and ACP1 on chromosome 2p

The genomic DNA-probe L2.30 was used to assign D2S1 to 2p23–pter by in situ hybridization. The RFLP revealed by Bgl II was then used for linkage studies in the Oslo-NHIK families segregating for the acid phosphatase ACPl protein polymorphism. Evidence for very close linkage was found by a lod score of +17.17 at recombination fraction θ = 001. By this close linkage 92 informative meioses could be inferred from the families and with only a single crossover. The upper probability limit to the recombination fraction is 006 according to the HGM 8 criterion. No association between ACP1 alleles and D2S1 Bgl II alleles was found. The Norwegian gene frequencies for 0251 were A1 (9–0 kb) = 065 and A2 (6.3 kb) = 0.35.     

HL-A Antigen, Gene, and Haplotype Frequencies in Denmark

Between 426 and 1,967 unrelated Danes have been HL-A typed for most presently known HL-A antigens of the LA (first), FOUR (second), and AJ (third) segregant series. Antigen, gene and haplotype frequencies with delta values are given. AJ series antigens are most strongly associated with some of the FOUR series antigens, and except for one case, the linkage disequilibrium between AJ and FOUR does not seem to be influenced by the LA series; the exception concerns HL-A9, RH-315, and 12: the RH-315 determinant is significantly more frequent on HL-A9, 12 haplotypes and on other HL-A12 carrying haplotypes. The term "superhaplotype" is suggested for gene constellations such as the HL-A9, RH-315, 12 "haplotype". It is suggested that the associations between cross-reacting antigens from one series with the same antigen from another series may reflect recent evolutionary divergence of the cross-reacting antigens.     

Linkage and association studies with C8A and C8B RFLPs on chromosome 1

Linkage relations for the C8A and C8B Bam HI RFLPs have been investigated. A peak lod score of 4·52 at recombination fraction zero was obtained between the two C8 genes. Combined with our previously obtained linkage data (Rogde et al. 1986) the maximum lod score is 7·53 at recombination fraction zero. The compiled C8-PGM1 linkage data from this and the previous study gave a maximum lod score of 22·02 at recombination fraction 0·11 (0·07–0·16) with no sex difference. A chromosome 1p reference marker, D1S57 , has been applied in this linkage study. A maximum lod score of 5·06 between the C8 cluster and D1S57 at θ= 0·18 (0·11–0·28) was recorded. The linkage analyses and triply informative families gave evidence that the C8 loci are situated about halfway between PGM1 and D1S57 on the short arm of chromosome 1. There was no evidence of allelic association between the C8A and C8B Bam HI RFLPs in 62 unrelated haplotypes.     

LINKAGE RELATIONSHIPS OF THE LOCI OF THE MAJOR HISTOCOMPATIBILITY COMPLEX IN FAMILIES WITH A RECOMBINATION IN THE HLA REGION

A number of families with an established recombination in the major histo-compatibility complex has been investigated for markers known to be coded by genes of this linkage group. The results provide further data on the relative position of the loci for HLA-A, HLA-B, HLA-C, HLA-D, Bf, Chido, Rodgers and PGM₃ on chromosome 6. A positive lodscore for linkage between HLA and blood group P was found; lodscores between HLA and nineteen other markers were negative.     

Posterior probability of linkage and maximal lod score

To detect linkage between a trait and a marker, Morton (1955) proposed to calculate the lod score z(θ₁) at a given value θ₁ of the recombination fraction. If z(θ₁) reaches +3 then linkage is concluded. However, in practice, lod scores are calculated for different values of the recombination fraction between 0 and 0·5 and the test is based on the maximum value of the lod score Zmax. The impact of this deviation of the test on the probability that in fact linkage does not exist, when linkage was concluded, is documented here. This posterior probability of no linkage can be derived by using Bayes' theorem. It is less than 5% when the lod score at a predetermined θ₁ is used for the test. But, for a Zmax of +3, we showed that it can reach 16·4%. Thus, considering a composite alternative hypothesis instead of a single one decreases the reliability of the test. The reliability decreases rapidly when Zmax is less than + 3. Given a Zmax of +2·5, there is a 33% chance that linkage does not exist. Moreover, the posterior probability depends not only on the value of Zmax but also jointly on the family structures and on the genetic model. For a given Zmax, the chance that linkage exists may then vary.     

C4 Chido 3 and 6 distinguish two diabetogenic haplotypes: HLA-B49, SC01,DR4,DQw8 and B8,SC01,DR3,DQw2.

The combination of the HLA complement allotypes BFS, C2C, C4AQ0 (deleted gene) and C4B1, termed SC01 complotype, usually present in the HLA-B8,DR3,DQw2 diabetogenic haplotype, has also been found in a novel "low frequency" HLA-B49,DR4,DQw8 haplotype associated with Spanish insulin-dependent diabetes mellitus (IDDM). Family studies of C4 antigenic determinants Rodgers/Chido and their specific C4d nucleotide sequences confirm that this novel haplotype bearing Chido -3, -6 is not due to a recent recombination from the common HLA-B8,DR3 haplotype bearing Chido 3,6; moreover, Chido analysis at the serological or DNA level is presently the only way to distinguish both SC01 complotypes, since BF, C2, steroid 21-hydroxylase and C4 genes do not reveal other differences by restriction fragment analysis. On the other hand, HLA-B49,SC01,DR4 is the first DR4-bearing IDDM-susceptible haplotype with a deleted C4 gene described so far and the only DR4-bearing haplotype found in the Spanish population. This report further supports the fact that extended haplotypes with deleted (or "not duplicated") genes in the class III region contain IDDM-susceptibility more often than non-deleted (or "duplicated") haplotypes in the Spanish and other Mediterranean populations.     

Rapid computer analysis of linkage data

A well tested and useful computer program for screening large amounts of pedigree data for linkage is described. All types of pedigree and genetic systems with up to six alleles present in the pedigree are allowed. An original procedure fills in uncertain genotypes and searches for internal inconsistencies in the data and is extremely thoroughgoing. Population gene frequencies are not used, although the program does take account of rare alleles. The scoring method is based on the division of the pedigree into sibships and the application of z-scores for the sexes considered separately wherever possible. About 2000 pedigrees have been analysed and scores for about the same number of pairs of loci are on tape. A mathematical exercise showed that the method of z-scores with score corrections, which is very much quicker than calculating complete likelihood, is acceptably efficient and its efficiency approaches 100% when scoring rare recessive systems. It is shown that the maximum value of the score is a good criterion for linkage, values of 3.1 and 4.0 indicating roughly 95% and 99% probabilities of linkage. The calculation of 2-unit support limits for the estimate of the recombination fraction between linked loci is advocated, and tables of these are given for syntenic loci. No new linkages were confirmed.     

Study of HL-A Genotypes in a Case of Familial Chronic Lymphocytic Leukaemia (CLL)

Blood groups and HL-A antigens were determined in 12 members of a family in which chronic lymphocytic leukaemia (CLL) occurs in four siblings. The same HL-A haplotype [Da25 (W30 + W31), HL-A13] was observed in the three diseased siblings tested.
This case, together with that of a similar family studied by Schweitzer et al. (1973), in which three out of five diseased siblings shared two antigens - HL-A2 and W5 - suggests a possible linkage between HL-A and susceptibility to CLL.     

A Note on the Asymptotic Properties of Affected-sib-pair Linkage Tests

This article concerns the asymptotic properties of linkage tests for affected‐sib‐pair data under the null hypothesis of no linkage. We consider a popular single‐locus analysis model where the unknown parameters are the disease allele frequency, the three penetrances for the three genotypes at the disease locus, and the recombination fraction between the marker locus and the disease locus. These parameters are completely confounded under the null hypothesis of no linkage. We show that 1) If the total variance of the trait (i.e., the additive variance plus the dominance variance) is “separated” from 0, then the likelihood ratio statistic has an asymptotic 0.5χ²₀+ 0.5χ²₁ distribution; 2) If the prevalence of the trait is “separated” from 0 and the recombination fraction is fixed at 0, then the likelihood ratio statistic has an asymptotic distribution which is a mixture of χ²₀, χ²₁ and χ²₂ . The implications of these results are discussed.     

Combined segregation and linkage analysis of nonsyndromic orofacial cleft in two candidate regions

We applied a complex segregation analysis to 46 pedigrees with a total of 121 nuclear families and 660 individuals, to verify hypotheses regarding the inheritance of OFC and linkage with markers on chromosomes 6 and 2. The pointer program for segregation analysis strongly rejected the hypothesis of no familial transmission of OFC in these families. When the hypothesis of a two-locus model was tested with comds , the analysis showed the presence of at least two loci and the model assuming a dominant major gene and a recessive modifier locus was statistically accepted. Given the fitted two-locus model, we tested for a possible linkage between the major OFC locus and the two markers studied. For D6S259, the estimate of the recombination fraction was θ=0.098, corresponding to a LOD score around 2.1. On the contrary, the data analysis concerning the D2S378 marker showed an estimate of the recombination fraction not significantly different from the independence hypothesis.     

Linkage analysis with misclassification at one locus

It is demonstrated that in linkage analyses with a misclassification error p of the phenotypes at one locus, a wrong specification of p leads to a bias in the estimate of the recombination fraction. In particular, if p is not taken into account at all in the analysis, the recombination fraction is overestimated. This is shown for offspring of phase-known double backcross matings. In the light of these results, the possible effects of dichotomization of quantitative phenotypes are discussed and illustrated with an example of the linkage between hypercholesterolemia and C3.     

HL-A8 AND HAPLOTYPE HL-A1-8 IN COELIAC DISEASE

HL-A antigens in fifty-four coeliac children and in parents and healthy siblings of forty-seven of these patients have been determined, thus allowing deduction of haplotypes and segregation analysis. HL-A8 frequency was found highly significantly increased (χ²= 37.92; P? 0.001) in coeliac children compared with a control group of 240 unrelated individuals; the increase of HL-A1 frequency is attributed to linkage disequilibrium between the genes involved. The haplotype HL-A1-8 frequency was also found significantly increased in coeliac children (χ²= 55.13; P? 0.001) with frequency elevation in the healthy siblings (χ²= 31.45; P? 0.001) in comparison with the controls. In segregation analysis a significant deviation from the expected distribution of HL-A8 (χ²= 15.38) and to a somewhat lesser degree of HL-A1 (χ²= 12.45 could be shown in coeliac children while there was no significant deviation in the distribution of these antigens detectable in their healthy siblings. Only a few haplotypes were frequent enough to be included in segregation analysis of haplotypes. Again in coeliac children a positive correlation with haplotype HL-A1-8 was found differing significantly from the expected distribution (χ²= 8.05), whereas this haplotype appeared almost evenly distributed in the healthy siblings of the patients.