Student Theses and Dissertations

Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

RU Laboratory

Friedman Laboratory


One of the current challenges in human genetics is to map genes for common, complex diseases. For powerful mapping of such phenotypes, the suggestions are to analyze the underlying quantitative traits with covariate corrections in large extended pedigrees with multipoint analysis. This is virtually impossible to do with the current linkage programs, due to the computation difficulties of exactly calculating every possibility. Instead, sampling methods that sample the most likely data configuration from all the possibilities need to be used. This has been implemented in the reversible jump Markov chain Monte Carlo method Loki, which can carry out segregation and linkage analysis on quantitative traits in large pedigrees with multipoint analysis. Loki can model the trait with covariates, identify the number of quantitative trait loci, linked loci, and estimate allele frequencies and gene effects. This method has a lot promise but has not been vigorously tested for complete genome scans. The first part of this study was to develop a strategy for carrying out genome scans using Loki and to evaluate the output. This was first done using the Genetic Analysis Workshop 12 simulated dataset with known answers. This resulted in a number of suggestions, such as initial single chromosome analysis, correction for polygenic effect, joint analysis of positive signals, and convergence analysis. Next these suggestions were applied to a real dataset from the population of Kosrae, the Federated States of Micronesia. This is a study of the population on the island of Kosrae, which one large extended pedigree and high prevalence of the common complex disorders that are known as Syndrome X: obesity, type II diabetes, hypertension, and dyslipidemia. This resulted in a number of additional suggestions, such as phenotypic and genotypic corrections, dealing with mixing issues, and inspection of L-graphs for signal reliability. Once this strategy was developed, the second part of this study was to use Loki to identify quantitative trait loci for the continuous traits associated with Syndrome X and stature. This resulted in quantitative trait loci for body mass index, hip circumference, weight, fasting blood sugar, systolic blood pressure, arterial blood pressure, apolipoprotein B, total cholesterol, and height. This also identified interesting chromosomal regions with slight signals for correlated traits on chromosomes 1, 2, 7, 9, 13, and 16. This study shows that Loki is a program that can powerfully and reliably carry out linkage analysis on quantitative traits that was previously impossible to do and finds loci for many of the quantitative traits related to common metabolic disorders as well as height.


A thesis submitted to the Faculty of The Rockefeller University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Included in

Life Sciences Commons