Date of Award

Spring 2020

Degree Type



Exercise Science

Director of Thesis

Dr. Mark Sarzynski

Second Reader

Jonathan Ruiz-Ramie, M.S.


Background. Obesity affects nearly a fifth of all children in the United States and is a common risk factor for numerous chronic diseases. Many risk factors contribute to obesity, including race and genetics. Obesity is partially heritable, and many genetic loci have been identified as being associated with body mass index (BMI) in adults. Some of these loci have also been associated with childhood BMI, with effects strengthening with age. A previous genome-wide association study (GWAS) identified five single nucleotide polymorphisms (SNPs) associated with changes in BMI in children aged 1 – 17 years.

Purpose. The purpose of the current study was to test whether these five SNPs were associated with BMI and other body composition measures in ninth graders from the Transitions and Activity Changes in Kids (TRACK) study population, in which 70% of participants identify as black.

Methods. DNA was extracted from cheek swabs of each TRACK participant, then used to determine the students’ genotypes for the five SNPs of interest. A genetic risk score was created based on the total number of risk (i.e., BMI-increasing) alleles carried across all five SNPs. General linear models were performed to examine associations between each SNP or the genetic risk score and multiple body composition measures, including BMI, fat mass, fat free mass (FFM), fat mass index (FMI), and percent body fat (PBF), controlling for age and sex, with another model additionally controlling for maturity.

Results. There were few associations between individual SNPs and the various body composition measurements. In whites, FAM120AOS SNP rs944990 was nominally (pFTOSNP rs1558902 was nominally associated with BMI, fat mass, FMI, and body weight, with associations strengthening with additional adjustment for maturity. The genetic risk score was nominally associated with BMI in whites, with the association being attenuated when maturity was added to the model. White students in the highest genetic risk score group (5 or more total risk alleles) had a significantly higher BMI than students in any other risk score group. The genetic risk score was not associated with BMI or other measures in black students.

Conclusions. We found limited evidence of an association between BMI-SNPs and a genetic risk score with body composition measurements in ninth graders. Studies with larger sample sizes are needed to allow for the identification of genetic loci connected with body composition in adolescents, particularly in blacks.

First Page


Last Page