Take home Exam #8 -- Due Wednesday, December 3rd
- Clustalx is a program that allows calculating multiple sequence alignments. What other things can you do with this program?
A) Calculate distance matrices (with or without correction for multiple substitutions), neighbor joining trees, and perform bootstrap analyses
B) Calculate distance matrices, parsimony trees, and perform bootstrap analyses
C) Calculate distance matrices, neighbor joining trees, do bootstrap analyses and evaluate user trees.
- Clustalx and clustalw perform progressive alignments and phylogentic reconstructions.
A) The guide tree is used to evaluate bootstrap samples.
B) The guide tree is calculated from pairwise alignments.
C) The guide tree is used to choose the order in which sequences are aligned into the multiple sequence alignment
D) The guide tree is stored as *.dnd file, and you tell clustal to use a user calculated tree as a guide tree.
E) B,C,and D are correct
- The lower the bootstrap value the better the support for the different parts of a phylogenetic tree. Correct/Incorrect
The above tree that was calculated in ClustalX from 1000 bootstrap replicates. It should be considered as unrooted; bootstrap support values are written to the right of the branch they pertain to. MYG denotes a myoglobin (oxygen buffering molecule in the muscle of animals); LGB2 denotes a plant globin that functions as an oxygen buffer in root nodules of leguminous plants; GLB5 PETMA denotes a hemoglobin (oxygen carrier in the blood) from a deep branching fish lineage; HBA and HBB denote the alpha and beta chains of the hemoglobin molecule, respectively.
The number 977 indicates that in 97.7% of the bootstrapped samples “HBA HUMAN” groups with “HBA HORSE”. Correct / incorrect
- In the above tree, the number 650 indicates that in 65.0% of the bootstrapped samples “MygPhyca” , “LGB2 LUPLU”, and “GLB5 PETMA” group together. Correct / incorrect
- In the above tree, the number 650 indicates that in 65.0% of the bootstrapped samples “HBA HUMAN”, “HBA HORSE”, “HBB HUMAN”, and “HBB HORSE” group together. Correct / incorrect
- Based on the above tree it is possible that GLB5 PETMA groups with (HBB Human, HBB Horse) in 35% of the analyses performed on the bootstrapped samples. Correct / incorrect
- In the MrBayes program, the sampling process explores
tree-space in a biased manner.
What measure does the program use to decide, if one tree is
“better” than the tree last examined?
- How does the heated chain in a Metropolis Coupled Monte
Carlo Markov Chain exploration differ from the chain that is not
heated?
- Why is it a good idea to include a heated chain in an
analysis?
Under which conditions are heated chains most useful? (Think about the hills and valleys in the likelihood landscape.)
- Assume that you have MrBayes estimating a parameter, i.e.
the magnitude of this parameter is changed during the biases random walk,
and its magnitude is considered in calculating the likelihood of each tree
visited. How could you
calculate the 90% credibility interval for this parameter?
- How does MrBayes approximate the posterior probability of
a bipartition (i.e. a branch in a tree)?
- What does the term “posterior probability” in the above
question denote?
A) the probability of the data given the model
B) the probability of an aspect of the model given the data
C) the probability of aspects of the model before one has looked at the data
- When calculating trees with clustalx/clustalw
you have the option to exclude all positions that have a gap in any of the positions
from the analyses. The default is to exclude gaps only from the pairwise alignments.
Under which conditions might it be advantages to turn on this option, under which
conditions might the default setting be preferable?
- How does protein parsimony treat gaps in an alignment by default?
- How is this different from the way clustalx treats gaps in phylogenetic reconstruction?
a) What tree does the dataset corresponding to point 1 support? How reliable is the support?
b) What tree does the dataset corresponding to point 2 support? How reliable is the support?
- What does the likelihood value of a tree refer to?
A) the probability of the data given the tree and a model of evolution
B) the probability of the tree given a model and the data
C) the maximized probability assuming that a given tree topology is true
- Pick the best answer that describes Maximum Likelihood analysis.
A) Finds the tree that explains sequence data with minimum number of substitutions.
B) Given a model for sequence evolution, find the tree under which the dataset has the highest probability.
C) Measures how much support for a partition is in your data given a method of analysis.
D) All of the above.
- How is does the program MrBayes assess posterior probabilities?
A) Through quartet puzzling
B) Through calculating Pi as Li/(L1+L2+L3), where Li is the maximum likelihood of tree with topology i
C) Through a biased random walk through tree and parameter space.
(Extra bonus question:)