ORIGINAL

Abstract

Animal models of aggression are an important tool for understanding what is correlated with differential aggression within a species. Epigenetics and its correlation to aggression is not an area that has thoroughly explored for gray wolves, Canis lupus. Through methylation data of 59 wolves, this project aimed to determine whether melanism has an effect on methylation, and if differential methylation is a significant factor in predicting aggression in wolves.

I synthesized and collected data that included aggressiveness ratings, wolf pedigree, sex, age, and pack information for predictor and covariate factors. Reduced Representation Bisulfite Sequencing (RRBS) was used to obtain the methylation data. Instead of focusing on the entire genome, I looked specifically for regions that are differentially methylated with respect to aggression. To accomplish this, I used linear mixed models to analyze the relationship between methylation and other external and environmental factors. After regions were identified, I determined their function, if they were silenced by melanism, and whether or not they play a significant role in regulating aggression. This method allowed me to assess the relative importance of factors such as sex, age, coat color, and methylation with respect to aggressive behavior.

The linear mixed model produced several hundred methylated sites, which were then ranked by p-value and grouped by position and beta values. This left three main region candidates to examine. Two of the groups were on chromosome 7, and both were associated with immune responses. A third group was on chromosome 10, and although non-coding was near protein coding regions. Although unable to determine a genetic mechanistic effect melanism has on a wolf’s genome, this project leads to further research on how methylated sites effect gray wolf aggression.