Investigation of genetic variation associated with complex diseases in time-series genome sequencing data with population genetics approaches

Genetic diseases are the products of evolutionary process. The demographic history of anatomically modern humans over the last 200,000 years has shaped the genetic structure associated with common diseases. It is possible to understand the dynamics of evolutionary forces affecting the genome in light of time-series genomic data. Ancient DNA provides time-series genomic information. With the genomic data generated in this way, it is possible to study the evolutionary forces affecting the genetic structure of common diseases. In this thesis, we investigated the temporal dynamics of genetic structure behind Obesity and Type 2 diabetes, two complex diseases with high prevalence worldwide over the last 10,000 years. A total of 319 ancient and 16 modern human genome sequencing data from Anatolia were analyzed for 18 Obesity-related, 415 Type 2 diabetes-related and 1000 neutral variants. Allele frequency trajectories were determined for all genetic variants using a maximum likelihood-based approach for three different periods: Neolithic (10326-6700 BC), post-Neolithic (7720-50 BC), and present-day. The genetic distance between ancient and modern Anatolian populations for variants associated with disease was calculated using FST and the results were compared with variants located in evolutionarily neutral regions. The periods in which allele frequencies and FST distributions differed significantly from neutral variants among the compared binary periods were Neolithic and Modern, Neolithic and Post-Neolithic (p<0.05, ANOVA). To test for positive selection, the population branch statistic (PBS) was calculated. As a result of the PBS, among the variants with selection signals, one variant was associated with Obesity and nine variants were associated with Type 2 diabetes with increasing allele frequencies from Neolithic to the present, and one variant was associated with Type 2 diabetes with increasing allele frequencies in Neolithic and post-Neolithic periods were discovered. Among these variants, TCF7L2, gene associated with rs35011184, is one of the genes associated with Type 2 diabetes with traces of selection. Anatolian ancient and modern populations were modelled under a neutral scenario using a forward-time genetic simulation tool. The allele frequency distributions calculated with the synthetic genome data were compared with the frequencies obtained from real data and the difference with the neutral model was tested. It was found that the increase in allele frequencies between the Neolithic and present-day was higher than the increase observed under the neutral scenario. It has been shown that the frequency of some of the alleles associated with Obesity and Type 2 diabetes has increased over the last 10,000 years, and that this change may imply for polygenic selection.