Faculty Bibliography

Trisomy 21 usually results from nondisjunction during meiosis I. In order to determine whether nondisjunction results from failure of normal chromosome pairing or premature unpairing, recombination frequencies were estimated between DNA polymorphic markers on the long arm of chromosome 21 in families containing one individual with trisomy 21. The recombination frequencies on chromosomes 21 that had undergone nondisjunction were then compared to those on chromosomes 21 that had disjoined normally. The data indicate that recombination is reduced between DNA markers on nondisjoined chromosomes 21. These results are consistent with the hypothesis that reduced chiasma formation predisposes to nondisjunction, resulting in trisomy 21 in humans.

The first portion of this communication reiterates the method of discarding single probands for studying the segregation ratio under incomplete ascertainment. The organization of the algebra leads to the second portion, in which the "self-contained" subsets are constructed with respect to both the segregation ratio and the ascertainment ratio. These subsets make a more detailed study of segregation and ascertainment possible.

A lod score method is provided for mapping genes relative to the centromere using family data from autosomal trisomies. Such gene-centromere mapping can be performed whenever two or more members of a meiotic tetrad can be recovered. The critical mapping parameter is not the recombination value theta or the map distance omega, but the probability of nonreduction in a heterozygous host, the probability of heterozygosity (nonreduction) is 1-gamma/2 for a meiosis I error and gamma for a meiosis II error. Under various assumptions regarding chiasma interference, gamma can be related to theta and omega. We provide specific methods for estimating gamma and theta from trisomy data using maximum likelihood, so that recombination may be studied on chromosomes that underwent nondisjunction.

Linkage analysis of neurofibromatosis was performed using genes on chromosomes 1, 8, 11, and 12. No linkage was found between NF and C-myc, AT 3, IGF-1, PTH, and gamma globin genes. Evidence for linkage was found between C-ets 1, on the long arm of chromosome 11 and NF in two families with a lod score of 1.88 at theta = 0. More families are being studied to confirm this linkage.

Initial and accurate inference of locus order and estimates of interlocus distances and interference can be obtained using seriation techniques. The analysis requires a matrix of recombination values that can be estimated by standard pairwise linkage analysis. This allows combination of results from individual investigators without reanalysis of basic pedigree material. Seriation can be performed without the use of a computer.

Seriation methods provide an accurate and efficient means of constructing preliminary multilocus genetic maps. By using both simulated and previously published empirical data, multipoint mapping by seriation was critically evaluated. Analysis of the simulated data sets showed that the seriation methodology could accurately estimate order and interlocus distances. Application to the empirical data demonstrated that seriation could obtain results directly comparable with those of other multipoint mapping methods. Techniques such as seriation can produce preliminary genetic maps that may be used as starting points for more computer-intensive maximum-likelihood multipoint techniques.

Trisomic individuals provide information for gene-centromere mapping, since two of the four chromatids in a meiotic tetrad can be recovered. When centromeric markers are available, linkage analysis between the centromere and any marker locus can be performed in nuclear families having one or more trisomic offspring. Since conventional linkage programs consider only disomic individuals, we have written a FORTRAN computer program, DSLINK, that performs gene-centromere linkage analysis on the basis of information on trisomic and disomic offspring. This program makes it possible to study the relationship between recombination and chromosome segregation.

When linkage between a recessive Mendelian disease and specific candidate genes is investigated, identity by descent scores in affected sib pairs may be used for tests of linkage and heterogeneity. Statistical tests for performing this analysis are presented. The efficiency and statistical power of the method are also investigated using computer simulations.

Forty-one multiplex families, from published sources and new data from the National Cancer Institute, segregating for Hodgkin's disease and HLA, have been studied. A reanalysis of these data strongly suggests a recessive mode of inheritance for susceptibility to Hodgkin's disease. The HLA haplotype sharing data between affected relatives demonstrate that approximately 60% of cases in multiplex families are due to an HLA-linked susceptibility gene, the remaining 40% being due to other familial factors. The data clearly support the hypothesis of etiological heterogeneity for Hodgkin's disease, with both HLA-linked and HLA-unlinked factors being responsible. Finally, there is an increased concordance of histological types between affected relatives, but this concordance seems independent of HLA sharing.