Biomedical subjects
Publications and source records attributed to O Mayo.
A complex disease may be defined as a disease that has a complex phenotype, no single causal agent and a range of influencing factors, both genetic and environment. The paper considers how the aetiology of such disorders may be unravelled using recently developed genetical and statistical methods. Problems are illustrated using ischaemic heart disease, congenital neural tube defects and depression as examples.
The DNA replication inhibitor peptide microcin B17 is shown to be a useful tool for preparing Escherichia coli maxicells. To illustrate its usefulness, we have identified polypeptides synthesized from pMccB17 and R100 IncFII miniplasmids. After comparing the respective polypeptides and the miniplasmid restriction maps, we concluded that these plasmids share extensive homology in the basic replicon but are different for an adjacent region (parD) that is involved in plasmid stability and maintenance.
The possible explanations for heterosis and heterozygous advantage have included the hypothesis that the metabolic versatility of heterozygotes for functional alleles of structural genes would enhance resistance to environmental insult, i.e. would result in enhanced developmental homeostasis. Evidence on this hypothesis is conflicting. The paper presents additional evidence, based on four human polymorphisms and 9 quantitative traits in a sample of mother-offspring data from Sweden. These data do not support the hypothesis of interest. Reasons for the conflicting results are discussed.
Both regression and correlation estimates of genetical variance and heritability for a quantitative trait influenced by a major gene can be obtained from the error variance-covariance matrix of MANOVA using relative-relative phenotype pairs as factors. The method is illustrated with parent-offspring data on red cell acid phosphatase phenotypes and serum acid phosphatase activity.
Associations between polymorphisms and disease are usually detected by comparing phenotype frequencies in affected individuals and controls, usually by the method of Woolf, which also allows assessment of heterogeneity between studies. The risk of reporting a chance spurious association could be reduced if family studies, such as sib comparisons, were carried out at the same time as the original survey, rather than after many surveys have been conducted.
Two recent methods for detecting major genes under continuous variation are investigated, by analysis of both simulated and real data. The major gene index of Karlin is shown to be sensitive to the distribution of gene effects in such a way that it may not detect major genes in certain cases. The intrafamily correlations of Matthysse et al. are shown to yield values difficult to reconcile with established patterns of inheritance for certain traits.
Major genes affecting quantitative traits are widespread and well known. However, major genes affecting unimodally distributed quantitative traits appear to be rare in populations of outbreeding organisms. This may largely reflect methods of analysis, or it may reflect the interaction of chance and selection on the frequencies of such major genes. If many quantitative traits have intermediate optima, or at least are subject to selection against extremes, it may be argued that selection will tend to eliminate segregation for genes of major effect more rapidly than for those of minor effect. This argument is tested by Monte Carlo methods. Results suggest that times to fixation do not differ greatly for major and minor genes determining a trait undergoing a moderate degree of centripetal selection in small populations but major genes are in general fixed first. The effect of linkage and the definition of 'large' and 'small' effects are discussed.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Divergence over time in flocks selected for clean fleece weight and for fecundity is measured by gene frequency changes at the R-r-i blood group, haemoglobin, and transferrin loci. No unequivocal effect of selection is demonstrable. Problems of interpretation even in flocks of known history are discussed.
The genetics of human family size is not readily susceptible to analysis. Some of the problems in its investigation are discussed, together with methods for overcoming them. In particular, the problem of obtaining a satisfactory model which will account for both the low heritability of family size and its over-dispersed distribution is examined in detail.
Heritability is well known to be a poor indicator of genetical variation in fitness. We show here that it is also an inadequate measure of genetical variation in fecundity.
The detection and estimation of the effects of paternal and maternal age and birth rank on mutation rate are considered. Smith's (1972) method and discriminant function techniques are compared using data on Apert's acrocephalosyndactyly and achondroplasia.
Under the same conditions which yield 1 + 4Nemu as the effective number of neutral alleles maintained by fresh mutation at a rate mu at an autosomal locus in a population of size Ne, the corresponding number at an X-linked locus is 1 + 3Nemu. The resultant bimodal frequency distribution for the effective number of alleles does not seem to have been used as the basis for analysis of data on actual numbers of alleles in natural populations.
The effect on sibship size distribution of the birth of a child with a genetical defect is considered for several different conditions. Family size continues to be over-dispersed in such cases, rather than showing any sign of reduced variation, though theoretical expectations about the correlation between numbers of normal and affected children are not well-supported by the data.
Explore the source record for details and available documents.
Explore the source record for details and available documents.