Genetic mapping of complex traits: the case of Type 1 diabetes.

Tutkimustuotos: OpinnäyteVäitöskirjaArtikkelikokoelma


The risk of developing Type 1 diabetes depends on the action of a number of genes in combination with environmental factors. The pathogenic process is unknown, but at the time of diagnosis the autoimmune destruction of the insulin producing beta cells of pancreas has proceeded to a level where an individual is unable to survive without insulin replacement therapy. Type 1 diabetes is the second most common chronic disease of children in Finland, with approximately 0.4% of total population being affected. The incidence is one of the highest in the world and is continuously increasing. The reasons for the upward trend in the incidence are not known. The main predisposing genetic region for Type 1 diabetes is the HLA locus on chromosome 6p21.3 (IDDM1). Recent genome scans have identified non-HLA loci linked to Type 1 diabetes, with much weaker effects than IDDM1. More accurate identification of the genes and of the DNA variants involved will lead to a better understanding of the disease.

The first part of this thesis is concerned with incidence trends: a survey of global incidence trends revealed strong evidence for a global increase. A hypothetical genetic explanation for the increase was given and studied, namely the possibility of non-Mendelian transmission of diabetes susceptibility alleles, increasing the pool of predisposing alleles in the population. The transmission probabilities of HLA A, B and DR alleles from parents to offspring were estimated from a nationwide Type 1 diabetes genetic epidemiological study carried out in Finland. Existence of strong non-Mendelian transmission could be ruled out, but minor deviations in the transmission probabilities may still be possible. A simple population model for the effects of modest transmission distortion on the incidence showed that this mechanism alone could not be the cause for the observed trend.

The second part of the thesis is concerned with the development of new methods for finding complex disease loci, with special applications to Type 1 diabetes. Data mining algorithms were used for linkage disequilibrium mapping (Haplotype Pattern Mining, HPM). The approach enables one to find loci even with strong allele and locus heterogeneity and associated low penetrances, which are expected in complex diseases, given that there are a sufficiently small number of founder disease alleles in the population. The method was applied to real data from Type 1 diabetic families from the UK, where the known susceptibility gene was accurately localized with less data than had previously been used by alternative methods. The presented studies demonstrate the advantages of utilizing the data mining approach in complex trait mapping.
Painoksen ISBN952-9528-70-1
Sähköinen ISBN952-9528-71-X
TilaJulkaistu - 2002
OKM-julkaisutyyppiG5 Tohtorinväitöskirja (artikkeli)


  • 118 Biotieteet

Siteeraa tätä