Refining Phylogenetic Analyses Phylogenetic Analysis of Morphological Data: Volume [, 1 ed.] 010619, 0106193, 97803674077, 97810374676, 97803678341

Cover
Half Title
Series
Title
Copyright
Contents
Preface
Author Biography
Abbreviations
Chapter 6 Summarizing and comparing phylogenetic trees
	6.1 Consensus methods
		6.1.1 Cluster-based methods
			6.1.1.1 Strict consensus trees
			6.1.1.2 Majority rule consensus trees
			6.1.1.3 Combinable components consensus
			6.1.1.4 Frequency difference consensus
		6.1.2 Methods not based on clusters
			6.1.2.1 Adams consensus
			6.1.2.2 Rough recovery consensus
			6.1.2.3 Median trees
	6.2 Taxonomic congruence vs. total evidence
	6.3 Pruned (=reduced) consensus and identification of unstable taxa
		6.3.1 Maximum agreement subtrees (MAST)
		6.3.2 Brute-force methods
		6.3.3 Triplet-based methods
		6.3.4 Improving majority rule or frequency difference consensus
		6.3.5 Swap and record moves
		6.3.6 Improving prune sets with an optimality criterion
	6.4 Zero-length branches and ambiguity
		6.4.1 Identification of zero-length branches and collapsing rules
		6.4.2 Consensus under different collapsing rules
		6.4.3 Numbers of trees, search effort
		6.4.4 Temporary collapsing
	6.5 Supertrees
		6.5.1 Semi-strict supertrees
		6.5.2 Matrix representation with parsimony (MRP)
		6.5.3 Other methods based on matrix representation
		6.5.4 Majority rule supertrees
	6.6 Anticonsensus
	6.7 Tree distances
		6.7.1 Robinson-Foulds distances (RF), and derivatives
		6.7.2 Group similarity (rough recovery)
		6.7.3 Rearrangement distances
		6.7.4 Distortion coefficient (DC)
		6.7.5 Triplets and quartets
	6.8 Implementation in TNT
		6.8.1 Consensus trees
		6.8.2 Temporary collapsing of zero-length branches, unshared taxa
		6.8.3 Tree comparisons and manipulations
		6.8.4 Identifying unstable taxa
		6.8.5 Supertrees
		6.8.6 Measures of tree distance
Chapter 7 Character weighting
	7.1 Generalities
	7.2 General arguments for weighting
		7.2.1 Homoplasy and reliability
	7.3 Successive approximations weighting (SAW)
		7.3.1 Weighting and functions of homoplasy
		7.3.2 Problems with SAW
		7.3.3 Potential solutions
	7.4 Implied weighting (IW)
		7.4.1 Weighting functions
			7.4.1.1 Weighting strength
			7.4.1.2 Maximization of weights and self-consistency
		7.4.2 Binary recoding, step-matrix characters
		7.4.3 Tree searches
		7.4.4 Prior weights
		7.4.5 IW and compatibility
	7.5 Weighting strength, sensitivity, and conservativeness
	7.6 Practical consequences of application of IW
	7.7 Problematic methods for evaluating data quality
		7.7.1 Tree-independent
		7.7.2 Probability-based
	7.8 Improvements to IW
		7.8.1 Influence of missing entries
		7.8.2 Uniform and average weighting of molecular partitions
		7.8.3 Self-weighted optimization and state transformations
		7.8.4 Weights changing in different branches
	7.9 Implied weights and likelihood
	7.10 To weight or not to weight, that is the question
		7.10.1 Criticisms of IW based on simulations
		7.10.2 Support and character reliability
		7.10.3 Weighting, predictivity, and stability
		7.10.4 Convergence between results of IW and equal weights
		7.10.5 Weighting in morphology vs molecules
	7.11 Implementation in TNT
		7.11.1 Self-weighted optimization
		7.11.2 Extended implied weighting
			7.11.2.1 Missing entries
			7.11.2.2 Uniform weighting of characters or sets
Chapter 8 Measuring degree of group support
	8.1 The difficulty of measuring group supports
	8.2 Bremer supports: definitions
		8.2.1 Variants of Bremer supports
			8.2.1.1 Relative Bremer supports (RBS)
			8.2.1.2 Combined Bremer supports
			8.2.1.3 Relative explanatory power
			8.2.1.4 Site concordance factors (sCF) and group supports
			8.2.1.5 Partitioned Bremer supports
	8.3 Bremer supports in practice
		8.3.1 Performing searches under reverse constraints
		8.3.2 Searching suboptimal trees
		8.3.3 Recording score differences during TBR branch-swapping
			8.3.3.1 The ALRT and aBayes methods
		8.3.4 Calculating average differences in length
	8.4 Resampling methods
		8.4.1 Plotting group supports
		8.4.2 Different resampling methods
			8.4.2.1 Bootstrapping
			8.4.2.2 Jackknifing
			8.4.2.3 Symmetric resampling
			8.4.2.4 No-zero-weight resampling
			8.4.2.5 Influence of number of pseudoreplicates
		8.4.3 Final summary of results
			8.4.3.1 Frequency-within-replicates (FWR) or strict consensus
			8.4.3.2 Frequency differences (GC) track support better than absolute frequencies
			8.4.3.3 A death blow to measuring support with resampling
			8.4.3.4 Frequency slopes
			8.4.3.5 Rough recovery of groups
		8.4.4 Search algorithms and group supports
			8.4.4.1 Search bias worsens the problems of saving a single tree
			8.4.4.2 Approximations for further speedups
			8.4.4.3 Worse search methods cannot produce better results
	8.5 Confidence and stability are related to support, but not the same thing
	8.6 Implementation in TNT
		8.6.1 Calculation of Bremer supports
			8.6.1.1 Searching suboptimal trees
			8.6.1.2 Searching with reverse constraints
			8.6.1.3 Estimation of Bremer supports via TBR
			8.6.1.4 Variants of Bremer support
		8.6.2 Resampling
			8.6.2.1 Options to determine how resampling is done
			8.6.2.2 Options to determine how results are summarized
			8.6.2.3 Tree searches
		8.6.3 Superposing labels on tree branches
		8.6.4 Wildcard taxa and supports
Chapter 9 Morphometric characters
	9.1 Continuous characters
		9.1.1 Ancestral states, explanation, and homology
		9.1.2 Heritability and the phylogenetic meaning of descriptive statistics
		9.1.3 Significant differences and methods for discretization
		9.1.4 Scaling and ratios
			9.1.4.1 Shifting scale using logarithms
			9.1.4.2 Ratios
		9.1.5 Squared changes “parsimony” and other models for continuous characters
	9.2 Geometric morphometrics
		9.2.1 Geometric morphometrics in a nutshell
			9.2.1.1 Superimposition and criteria for measuring shape differences
			9.2.1.2 Symmetries
		9.2.2 Problematic proposals to extract characters from landmarks
		9.2.3 Application of the parsimony criterion: phylogenetic morphometrics
		9.2.4 Shape optimization in more detail
			9.2.4.1 Fermat points and iterative refinement of point positions
			9.2.4.2 Using grid templates for better point estimates
			9.2.4.3 Missing entries and inapplicable characters
		9.2.5 Landmark dependencies, scaling
		9.2.6 Implied weighting and minimum possible scores
			9.2.6.1 Weighting landmarks or configurations
			9.2.6.2 The minimum (∑Smin) may not be achievable on any tree
		9.2.7 Ambiguity in landmark positions
			9.2.7.1 Coherence in reconstructions of different landmarks
		9.2.8 Dynamic alignment of landmarks
		9.2.9 Other criteria for aligning or inferring ancestral positions
			9.2.9.1 Least squares or linear changes
	9.3 Choice of method and correctness of results
	9.4 Implementation in TNT
		9.4.1 Continuous (and meristic) characters
		9.4.2 Phylogenetic morphometrics
			9.4.2.1 Reading and exporting data
			9.4.2.2 Alignment
			9.4.2.3 Scoring trees, displaying, and saving mapped configurations
			9.4.2.4 Settings for estimating coordinates of landmark points
			9.4.2.5 Weights, factors, minima
			9.4.2.6 Group supports
			9.4.2.7 Ambiguity
Chapter 10 Scripting: The next level of TNT mastery
	10.1 Basic description of TNT language
	10.2 The elements of TNT language in depth
		10.2.1 Getting help
		10.2.2 Expressions and operators
		10.2.3 Flow control
			10.2.3.1 Decisions
			10.2.3.2 Loops
		10.2.4 Arguments
		10.2.5 Internal variables
		10.2.6 User variables
			10.2.6.1 Declaration
			10.2.6.2 Assignment
			10.2.6.3 Access
		10.2.7 Efficiency and memory management
	10.3 Other facilities of the TNT language
		10.3.1 Goto
			10.3.1.1 Handling errors and interruptions
		10.3.2 Progress reports
		10.3.3 Handling input files
		10.3.4 Formatted output
			10.3.4.1 Handling strings
		10.3.5 Arrays into and from tables
		10.3.6 Automatic input redirection
		10.3.7 Dialogs
		10.3.8 Editing trees and branch labels
		10.3.9 Tree searching and traversals
		10.3.10 Most parsimonious reconstructions (MPRs)
		10.3.11 Random numbers and lists, combinations, permutations
	10.4 Graphics and correlation
		10.4.1 Plotting graphic trees
		10.4.2 Bar plots
			10.4.2.1 Heat maps
		10.4.3 Correlation
		10.4.4 Scatter plots
	10.5 Simulating and modifying data
	10.6 A digression: the C interpreter of TNT
	10.7 Some general advice on how to write scripts
References
Index