Document Type

Dissertation - Open Access

Award Date


Degree Name

Doctor of Philosophy (PhD)

Department / School

Mathematics and Statistics

First Advisor

Gemeschis Djira


Breast cancer is the second most fatal cancer in the world and one of the most highly harmful cancers from which people suffer. Breast cancer studies have been able to uncover some knowledge about genetic susceptibility for familial breast cancer in humans. Hence, determining genetic factors may potentially help track the disease, as well as discover the cancer in early stages, or perhaps before it starts. In addition, this may allow early determination of possible treatment strategies which will make it easier to prevent the disease. In this context, it is important to determine whether the heritability of breast cancer incidence is greater than zero, which can be investigated if there is a potential genetic component playing a role in the incidence of the disease. Traits with zero heritability are said to be completely subject to environmental factors, so genetics has no effect at all. Heritability is important because it indicates the extent of genetic variations which could provide a reason for the infection. In the case that heritability is found to be greater than zero, it is useful to estimate the single nucleotide polymorphism (SNP) effects, which may potentially determine the genes or the genomic regions that are associated with the incidence of breast cancer. This study used data for three families with BRCAx as exome sequences provided by the University of Nebraska Medical Center and the Institutional Review Boards of Creighton University. Specifically, the data consisted of pedigree information for 167 individuals from three families, including information on whether each person had breast cancer or not (binary trait, positive or negative). Genomic data was available for 22 individuals among the 167. Theoretically, heritability as well as SNP effects can be estimated using a variety of approaches, but given the data available for this study, the best strategy was to combine both the pedigree-based data and the genomic data in one matrix. This matrix offers an advantage over other approaches that use only one of these datasets. The data was analyzed using a threshold model and Gibbs sampling algorithm to estimate the heritability of breast cancer incidence, as well as to predict SNP effects. The binary response variable for breast cancer incidence was modeled such that gender (2 levels) and family (3 levels) were the fixed effects. The effect of the subjects was the only random effect in the model. The heritability estimate was approximately 28%, indicating that there is a considerable genetic component underlying the incidence of breast cancer. In addition, the Genome-Wide Association Study (GWAS) analysis revealed that breast cancer is a complex trait, possibly controlled by many genes. However, some areas on the genome (specifically, chromosomes 1, 2, 4, 8, 14 and 16) may include candidate genes associated with breast cancer incidence. These genes might be responsible for this type of cancer and play important roles in susceptibility for the disease. The 20 SNPs with highest effects explained more than 3.5 % of the genetic variance, which is a good indicator that their genes are associated with breast cancer. The results of this study open the door for more research on breast cancer incidence. Despite the limitations related to the small sample used, the results of this study could be considered a first step for future work and investigation. Further studies using larger data sets may reveal more information on this complex trait.

Library of Congress Subject Headings

Breast -- Cancer -- Genetic aspects.
Familial diseases.
Gene mapping.


Includes bibliographical references (83-91)



Number of Pages



South Dakota State University



Rights Statement

In Copyright