I conducted an internship at the Texas Biomedical Research Institute in the spring of 2013. My internship supervisor, Dr. Anthony Comuzzie, is currently working on projects that study the genetics of coronary artery disease in Alaskan Natives, the genetics of atherosclerosis in Mexican Americans, and a study on diet and genotype in primate atherosclerosis.
The San Antonio Family Heart Study (SAFHS) focuses on the genetics of atherosclerosis in Mexican-Americans. Despite the fact that heart disease is the number one cause of death among Mexican-Americans, SAFHS is the only population-based genetics study in Mexican-Americans. Atherosclerosis (heart disease) is associated with many risk factors such as obesity, glucose intolerance, and hypertension. SAFHS tests the hypothesis that the genes that influence variation in these risk factors are likely to predispose individuals to atherosclerosis (Mitchell et. al. 1996). So far SAFHS has identified genes associated with insulin, fat mass and many others (Guowen et al 2004, Comuzzie et al. 2007). In the future, the scientists of Texas Biomed hope to be able to identify people with at-risk genes in order to tailor treatment to their specific genotype.
My work on the San Antonio Family Heart Study began with data management. Before I could conduct any statistical analyses on the dataset from SAFHS, I had to clean-up or adjust the data so that it resembled a normal distribution curve. In this case, the data consisted of various anthropometric measurements associated with obesity (BMI, fat mass, waist circumference), glucose, insulin, and levels of inflammatory cytokines taken from blood tests. I used PEDSYS, a UNIX based software package to remove the outliers so that the data fell within a normal distribution.
Next, I had to adjust the data so that it was compatible with SOLAR. SOLAR is also a UNIX based software package but it was designed for complex statistical analysis. In order for SOLAR to display data properly, rows and columns of a dataset can only be a certain number of characters long. If I made any changes to the variables of a data set, I had to make sure that these changes were compatible with SOLAR. This process is usually more time consuming than the statistical analysis.cOnce I cleaned up the data and made sure that the dataset was compatible with SOLAR, I began loading files necessary for statistical analysis.
I began my analysis of the SAFHS data by compiling basic descriptive statistics. I used the PEDSYS tally command on my phenotype file to obtain the means of obesity related measurements such as BMI, fat mass, fat free mass, and waist circumference grouped by sex. PEDSYS automatically performed calculations and I compiled the results in data tables. .
Next, I loaded my phenotype file into SOLAR, which allowed me to calculate heritabilities for my traits of interest. Heritability (h2), or narrow-sense heritability, is the proportion of total phenotypic variance that is due to additive genetic effects. Additive genetic variation is the variation that is the result of multiple genes contributing a small effect on the phenotype. Narrow-sense heritability requires a pedigree study because it uses measurements from both parents and offspring to measure the proportion of phenotypic variance that is the result of genetic factors. SOLAR is able to differentiate these values based on the pedigree file, and I simply specified the trait I wanted heritability for and typed in the command that adjusts for the variance due to sex and age (Figure 2). Heritability always ranges from zero to one.
My internship at Texas Biomed allowed me to see how scientific research is conducted across multiple fields such as genetics, epidemiology, and anthropology. I learned that identifying the genetic causes of disease is a complex and extensive process that often takes years to complete. This internship allowed me to utilize my knowledge of genetics and better understand the usage of statistics in population studies focusing on ethnic minorities and their cultures.
My internship at Texas Biomed has encouraged me to pursue graduate studies in epidemiology and medical anthropology. I plan to attend graduate school in the near future. An internship at Texas Biomedical Research Institute provides an opportunity to participate in the research process and learn how cultural aspects affect epidemiology.