作者： Andrea Benazzo
DOI:
关键词:
摘要: The main goal of population genetics is to understand the factors that affect genetic variation within a species. Mathematical models are used predict effects on processes such as mutation, recombination, selection, migration and size changes, but analytical results difficult obtain when these interact equilibrium conditions not met. In situations, common in real biological systems especially recent human activities (e.g., stocking, urbanization, overhunting) perturb natural populations, computer simulations can be very useful. A simulation virtual experiment which model mimic process study its properties. It an excellent tool for understanding functioning complex systems. Simulations generally make predictions about validate statistical methods, properties different sampling strategies, estimate parameters from data. In this thesis, I applied address questions intractable with other methods. First, analyzed violating assumption panmixiamade by “Extende Bayesian Skyline Plot” (EBSP) method. showed influence inferred demographic history population, suggesting wrong dynamics. Second, analyse performance EBSP method reconstructing decline compare schemes proportions modern ancient DNA. identified some scheme clearly positively reconstruction, providing simple hints planning both samples available. Third, familiarized “Approximated Computation” methodology contributed review article presenting features, pros cons, approach. Fourth, ABC procedure analyze hybridization genus Chionodraco, evaluate power context. Realistic were defined compared, evidence was found occurred only interglacial periods. Taken together, presented thesis confirm importance evolutionary biology. If we consider increasing availability packages, along speed storage capacity personal computers clusters, it easy genomic data will spread many fields better explore more realistic, consequently complex, models.