Lecture 20: Phylogenetics part 3: Tree building Flashcards
Phenetics (5)
-Groups taxa based on overall similarity
-Compares as many characters as possible
-no phylogenetic assumptions made
-Assumes that the contribution of homoplasy to overall similarity will be outweighed by the contribution of homology if enough characters are compared
-Critics argue this is not a reliable method
Neighbor-joining method (4)
-Represents a phenetic approach
-Clusters taxa so that the most similar forms are grouped together
-Many possible trees but ones with minimal distance among total taxa are preferred
-Fast but not the most accurate = a good starting point for more accurate methods like cladistics
Cladistics (6)
-Assumes that degree of relatedness does not equal degree of similarity
-Classifies orgs according to order and branches arose along a dichotomous tree
-Parsimony analysis of synapomorphies determines which phylogeny would require the fewest changes to illustrate evolutionary relationships
-therefore only shared, derived characters (synapomorphies) are informative
-Parsimony informative characters are shared by two or more taxa (ex exothermic shared between reptiles and amphibians)
-Characters unique to one lineage are uninformative, but may contribute to branch length / number of changes (ex hair in mammals)
Parsimony analysis of synapomorphies
Way of making a tree based on synapomorphies, creating the simplest tree (one with the least amount of required changes)
Evaluating trees (4)
-Done statistically using a method called “bootstrapping”
-Builds replicate trees by creating new data sets from the existing one through repeated sampling, resulting in a new sequence that is a random subset of the original
-Ex if there are 300 sites on a sequence, it will randomly select one as the new starting point and build a tree around it, then selects another and so on.
-Generates a consensus tree indicating in what percentage of the trees each branch occurs in
Bootstrapping rules of thumb (3)
-Values of 70+ suggest that particular grouping is reliable
-Values of 50-70 should be considered suggestive not conclusive
-Values bellow 50 suggest that particular grouping is not strongly supported by the dataset
Molecular clocks (3)
- Question: Do molecular traits evolve in a clock-like fashion (ie at a set rate)?
-Several attempts have been made to calibrate this “molecular clock”
-could be used to measure genetic differences between two taxa whos divergence is known from fossil / geographical records
mtDNA and molecular clocks (2)
-When examining changes in mitochondrial genome:
-In 1979, brown estimate the rate of this clock to be about 2% in mammals using fossil records
-Later, it was estimated that the same rate ( ~2%) was occurring in sea urchins, then in butterflies and geese
Molecular clock example (3)
-In snapping shrimp there are 7 morphological species pairs, each with one member on either side of Isthmus of Panama
-The clock showed that the genetic distance between sister species was not identical between pairs
-Lead to the conclusion that each pair isolated independently over a gradual rate, starting with the deep water species
Issues to the Molecular clock (2)
-Suggestions that mutation rate is faster in organisms with shorter generation times as well as differences in rate between cold and warm blooded species
-calibration using fossil and geological records only can estimate actual divergence dates, meaning that the start date of the “clock” is unknown
WHIPPO phylogeny (2)
-WHIPPO = whale + Hippo
- relationship between cetaceans (whale, dolphins, porpoises) and ungulates (horses, zebras, hippos, sheep etc) suggested by skeletal characters
-Proposed as sister group to artiodactyls (cows, deer, hippos, pigs, camels) with horses and rhinos as an outgroup because they have odd-toes
WHIPPO hypothesis (4)
-Suggests that the traits of whales and hippos that were thought to be convergent adaptations for aquatic life might actually be synapomorphies
-Morphological data places hippos as sister group to deer and cows, but molecular data places hippos as a sister group to whales
-fossil data showed that artiodactyls were descendants of a common ancestors from pulley-shaped structures in their ankle called astragalus BUT later on this structure was also found in fossilized whales who had hind limbs
-Additionally they found highly reliable (complex) molecular data in the form of SINEs and LINEs at the exact same locations in whales and hippos. Meaning that convergence was highly unlikely
