The lecture on Molecular Evolution and Phylogenetics, delivered by Ardavan F. Oskooi on October 8, 2008, covers the motivations, methods, and algorithms for constructing phylogenetic trees. The primary focus is on using DNA sequence data to determine evolutionary relationships among species, with an emphasis on detecting evidence of natural selection. Key topics include: 1. **Motivations**: Evolutionary divergence through natural selection, hybridization, and extinction. The shift from traditional phylogeny based on physiological data to genomic data for more accurate relationships. 2. **Probabilistic Models**: Jukes-Cantor and Kimura models for modeling evolutionary divergence. 3. **Tree Types**: Cladograms, phylograms, and ultrametric trees, each with different interpretations of branch lengths. 4. **Tree Building Algorithms**: - **UPGMA**: A simple hierarchical clustering method that produces ultrametric trees. - **Neighbor-Joining (N-J)**: A more sophisticated method that generates phylogram trees with branch lengths proportional to evolutionary rates. - **Parsimony**: A method that minimizes the number of substitutions to find the most parsimonious tree, using dynamic programming and traceback. - **Maximum Likelihood**: Combines statistical models with observed data to predict various evolutionary features, though it is more complex. The lecture also discusses the challenges and considerations in inferring phylogenies, such as the distinction between gene and species divergence, the impact of back mutations, and the importance of accurate distance matrices.The lecture on Molecular Evolution and Phylogenetics, delivered by Ardavan F. Oskooi on October 8, 2008, covers the motivations, methods, and algorithms for constructing phylogenetic trees. The primary focus is on using DNA sequence data to determine evolutionary relationships among species, with an emphasis on detecting evidence of natural selection. Key topics include: 1. **Motivations**: Evolutionary divergence through natural selection, hybridization, and extinction. The shift from traditional phylogeny based on physiological data to genomic data for more accurate relationships. 2. **Probabilistic Models**: Jukes-Cantor and Kimura models for modeling evolutionary divergence. 3. **Tree Types**: Cladograms, phylograms, and ultrametric trees, each with different interpretations of branch lengths. 4. **Tree Building Algorithms**: - **UPGMA**: A simple hierarchical clustering method that produces ultrametric trees. - **Neighbor-Joining (N-J)**: A more sophisticated method that generates phylogram trees with branch lengths proportional to evolutionary rates. - **Parsimony**: A method that minimizes the number of substitutions to find the most parsimonious tree, using dynamic programming and traceback. - **Maximum Likelihood**: Combines statistical models with observed data to predict various evolutionary features, though it is more complex. The lecture also discusses the challenges and considerations in inferring phylogenies, such as the distinction between gene and species divergence, the impact of back mutations, and the importance of accurate distance matrices.