Search PubMed⌕ Search

Biomedical subjects

D E Weeks

Publications and source records attributed to D E Weeks.

At least 19 recordsLinked to original sources
PubMed1999

An optimal algorithm for automatic genotype elimination.

In an effort to accelerate likelihood computations on pedigrees, Lange and Goradia defined a genotype-elimination algorithm that aims to identify those genotypes that need not be considered during the likelihood computation. For pedigrees without loops, they showed that their algorithm was optimal, in the sense that it identified all genotypes that lead to a Mendelian inconsistency. Their algorithm, however, is not optimal for pedigrees with loops, which continue to pose daunting computational challenges. We present here a simple extension of the Lange-Goradia algorithm that we prove is optimal on pedigrees with loops, and we give examples of how our new algorithm can be used to detect genotyping errors. We also introduce a more efficient and faster algorithm for carrying out the fundamental step in the Lange-Goradia algorithm-namely, genotype elimination within a nuclear family. Finally, we improve a common algorithm for computing the likelihood of a pedigree with multiple loops. This algorithm breaks each loop by duplicating a person in that loop and then carrying out a separate likelihood calculation for each vector of possible genotypes of the loop breakers. This algorithm, however, does unnecessary computations when the loop-breaker vector is inconsistent. In this paper we present a new recursive loop breaker-elimination algorithm that solves this problem and illustrate its effectiveness on a pedigree with six loops.

Algorithms↗
PubMed1992

A multilocus extension of the affected-pedigree-member method of linkage analysis.

The affected-pedigree-member (APM) method of linkage analysis is designed to detect departures from independent segregation of disease and marker phenotypes. The underlying statistic of the APM method operates on the identity-by-state relations implied by the marker phenotypes of the affected within a pedigree. Here we generalize the APM statistic to multiple linked markers. This generalization relies on recursive computation of two-locus kinship coefficients by an algorithm of Thompson. The distributional properties of the extended APM statistic are investigated theoretically and by simulation in the context of one real and one artificial data set. In both examples, the multilocus statistic tends to reject, more strongly than the single-locus statistics do, the null hypothesis of independent segregation between the disease locus and the marker loci.

Genetic Linkage↗
PubMed1991

Genetic mapping of "Lubag" (X-linked dystonia-parkinsonism) in a Filipino kindred to the pericentromeric region of the X chromosome.

"Lubag" is an X-linked disorder causing dystonia and parkinsonism that has only been described in families from the Philippines, principally from the island of Panay. We have established linkage between the disease phenotype "lubag" and DNA markers which span the Xp11.22-Xq21.3 region by using a large Filipino family with 8 affected men in three generations. These DNA markers define an interval of about 20 centimorgans in the pericentromeric region of the X chromosome as the most likely site of the disease locus XDPD (X-linked dystonia-parkinsonism). XDPD has a maximum multipoint log likelihood ratio score (Zmax) of about 4.6 over the interval from Xq12 to Xq21.31 (DXS159-DXYS1X). The co-occurrence of dystonia and parkinsonism in lubag and in other known disorders suggests there may be a common pathogenetic mechanism. Identification of the genetic defect in this family may provide an important clue toward understanding the pathogenesis and pathophysiology of both dystonia and parkinsonism.

Adult↗
PubMed1991

Assessment of chronic gamma radiosensitivity as an in vitro assay for heterozygote identification of ataxia-telangiectasia.

Ataxia-telangiectasia (A-T) is a rare human autosomal recessive disorder characterized by, among other symptoms, catastrophic reaction to conventional radiotherapy. A-T heterozygotes are clinically asymptomatic and their fibroblasts are intermediate in radiosensitivity between homozygotes and normals. We have attempted to identify heterozygotes by assaying for cellular hypersensitivity to chronic gamma irradiation. Cultured dermal fibroblast strains from 13 control subjects and 55 members from a large Amish pedigree segregating for A-T were assayed for loss of colony-forming ability (CFA) in response to 137Cs gamma radiation delivered at a dose rate of 0.8 cGy/min. For each strain, multiple dose-response curves were summarized in a composite D10 value (dose, in cGy, reducing colony survival to 10%). The D10's of the clinically normal controls and of those pedigree members with known A-T genotype formed a trimodal distribution, with the seven obligate heterozygotes displaying an average value (516 cGy) intermediate between that of the 10 healthy controls (797 cGy) and that of the two affected patients (154 cGy). The D10's were modeled statistically using Gaussian penetrance functions. The most parsimonious model yielded a significant difference in D10 means for heterozygotes and normal homozygotes, a significant donor age effect, but no sex effect. We compared probabilistic identification of heterozygotes based on D10 values with identification based on linkage data for two markers, THY1 and D11S144, closely linked to the A-T gene. This comparison revealed that the D10 data were appreciably less informative than the linked markers. Indeed, the extensive overlap between D10 values for heterozygotes and normal homozygotes precludes the use of postirradiation CFA for either accurate identification of heterozygotes or chromosomal mapping of the A-T gene.

Ataxia Telangiectasia↗
PubMed1990

Genetic epidemiology of bilateral breast cancer: a linkage analysis using the affected-pedigree-member method.

We used the affected-pedigree-member (APM) method to conduct linkage analyses on 19 pedigrees in which the probands had premenopausal bilateral breast cancer. This method analyzes all affected pairs of relatives, as opposed to siblings only, and incorporates into the analyses information on the frequency of marker alleles. Fourteen codominant marker systems were evaluated in two separate analyses. In the first, only premenopausal cases of breast cancer were coded as affected because we assumed that postmenopausal cases were due to a different etiology. In the second analysis, all cases of breast cancer were coded as affected, irrespective of menopausal status. In the premenopausal-cases-only analysis, we observed evidence suggestive of nonindependent segregation for C3 and ESD. In the all-cases analysis, we observed much weaker evidence for C3 and ESD and noted a suggestion of nonindependent segregation for AMY2 and PGM1.

Adult↗
PubMed1990

Measuring the inflation of the lod score due to its maximization over model parameter values in human linkage analysis.

A computer-simulation method is presented for determining and correcting for the effect of maximizing the lod score over disease definitions, penetrance values, and perhaps other model parameters. The method consists of simulating the complete analysis using marker genotypes randomly generated under the assumption of free recombination. It is applicable as a "post-treatment" to linkage analyses of any trait with an uncertain mode of inheritance and/or disease definition. When the method is applied to a linkage analysis of schizophrenia versus chromosome 5 markers, we find that, in this specific case, the P-value associated with a maximum lod score of 3 is equal to 0.0003. We also find that a lod score of 3.0 should be "deflated" by approximately 0.3 to 1 units, and, by tentative extrapolation, the observed lod score of 6.5 should be "deflated" by 0.7 to 1.5 units.

Chromosomes, Human, Pair 5↗
PubMed1990

A primary linkage map of the human chromosome 11q22-23 region.

We have constructed a genetic map of the human chromosomal region 11q22-23 by multipoint linkage analysis of 13 DNA polymorphisms that we have condensed into eight loci. An analysis for linkage disequilibrium between tightly linked probe/enzyme systems allows us to make specific recommendations for future DNA typing at these loci. The resulting sex-averaged multipoint map spans approximately 80 cM and differs considerably from previously reported genetic maps of this region. Our mathematically derived "most likely order" of the markers is compatible with physical mapping data using somatic cell hybrids. The known localizations of at least 14 functional genes and several disease loci to 11q22-23, including ataxia telangiectasia, make the mapping of this region especially relevant to studies of disease pathogenesis.

Chromosomes, Human, Pair 11↗
PubMed1990

Report of a workshop on genetic linkage studies in schizophrenia.

A workshop on genetic linkage studies in schizophrenia was held at Columbia University's Arden House Conference Center in October 1989. This report summarizes the contents of invited talks by Drs. Arno Motulsky and T. Conrad Gilliam and the discussions at the five workshop sessions. Topics of the workshop sessions were (1) diagnostic boundaries and hierarchies in schizophrenia, (2) genetic models and linkage parameters, (3) selection and ascertainment of pedigrees, (4) future extensions of molecular genetics strategies, and (5) possibilities for future collaboration.

Genetic Linkage↗
PubMed1989

Trials, tribulations, and triumphs of the EM algorithm in pedigree analysis.

The EM algorithm is an iterative method for finding maximum-likelihood estimates. Its advantages often include numerical stability, simplicity of computer implementation, and natural incorporation of parameter constraints. However, the EM algorithm must be tailored to each specific problem. Smith (1957) and Ott (1977, 1979) have accomplished this for a variety of problems in human pedigree analysis. The present paper clarifies their theory by presenting it from a modern perspective. Five practical numerical examples are also given in an attempt to assess the value of the EM algorithm in realistic genetic modelling. These examples deal with racial admixture, linkage homogeneity, classical segregation analysis, a Mendelian latent trait model for schizophrenia, and a heterozygote detection assay for Ataxia-telangiectasia. Comparison with a quasi-Newton method of optimization reveals that the EM algorithm generally converges more slowly, but also more stably.

Algorithms↗
PubMed1989

Efficient computation of lod scores: genotype elimination, genotype redefinition, and hybrid maximum likelihood algorithms.

Calculation of multilocus lod scores presents challenging problems in numerical analysis, combinatories, programming, and genetics. It is possible to accelerate these computations by exploiting the simple pedigree structure of a CEPH-type pedigree consisting of a nuclear family plus all four grandparents. Lathrop et al. (1986) have done this by introducing likelihood factorization and transformation rules and Lander & Green (1987) by the method of 'hidden Markov chains'. The present paper explores an alternative approach based on genotype redefinition in the grandparents and systematic phase elimination in all pedigree members. All three approaches accelerate the computation of a single likelihood. Equally relevant to multilocus mapping are search strategies for finding the maximum likelihood estimates of recombination fractions. Hybrid algorithms that start with the EM algorithm and switch midway to quasi-Newton algorithms show promise. These issues are investigated in the context of a simulated 10 locus example. This same example allows us to illustrate a simple strategy for determining locus order.

Algorithms↗
PubMed1988

The affected-pedigree-member method of linkage analysis.

This paper describes a generalization of the affected-sib-pair method of linkage analysis to pedigrees. By substituting identity-by-state relations for identity-by-descent relations, we develop a test statistic for detecting departures from independent segregation of disease and marker phenotypes. The statistic is based on the marker phenotypes of affected pedigree members only. Since it is more striking for distantly affected relatives to share a rare marker allele than a common marker allele, the statistic also includes a weighting factor based on allele frequency. The distributional properties of the statistic are investigated theoretically and by simulation. Part of the theoretical treatment entails generalizing Karigl's multiple-person kinship coefficients. When the test statistic is applied to pedigree data on Huntington disease, the null hypothesis of independent segregation between the marker locus and the disease locus is firmly rejected. In this case, as expected, there is a loss of power when compared with standard lod-score analysis. However, our statistic possesses the advantage of requiring no explicit assumptions about the mode of inheritance of the disease. This point is illustrated by application of the test statistic to data on rheumatoid arthritis.

Female↗
PubMed1987

Preliminary ranking procedures for multilocus ordering.

N linked loci can be arranged in N!/2 possible orders. We describe two criteria for providing a preliminary ranking of the possible orders based on the N(N-1)/2 pairwise lod score curves for the loci. For a given order the first criterion is the sum of the N-1 maximal lod scores corresponding to the adjacent pairs of loci in the order. The second criterion is the minimum of a least-squares problem due to J.M. Lalouel (1977, Heredity 38(1): 61-77). This least-squares problem requires the maximum likelihood recombination fraction estimates and their standard errors. For N small it is feasible to evaluate these measures for every possible order. For N large we use a simulated annealing algorithm. This gives a fairly complete listing of the best-candidate orders without sampling every possible order. These ranking methods are applied to data from linkage groups on chromosomes 1, 6, 11, and 13.

Algorithms↗
PubMed1987

Genetic linkage studies of ataxia-telangiectasia: phenotypic blood markers.

We have tested thirty-two phenotypic blood markers on sixteen families with with ataxia-telangiectasia (AT) in an attempt to identify the chromosomal location of the AT gene(s). Although at least five complementation groups have been defined, it is not known whether the corresponding AT genes are clustered or dispersed in the genome. Both clustered and dispersed genetic models were considered in linkage analyses. No significant linkages were found. The data exclude approximately 7 per cent of the autosomal genome for a 'clustered' model and 2 per cent of the autosomal genome for a 'dispersed' model. Several genomic areas were identified which warrant further study.

Ataxia Telangiectasia↗