Return to Main Index page

I. Phylogenetic analyses: many algorithms, a few core principles

Goal: arrangement of taxa on a "tree of life"
            understand the evolution of characters;
            process is best understood when pattern is clear --
                that is, many evolutionary questions require a well-resolved phylogeny
Gene trees vs. species trees
Nuclear vs. organellar (mtDNA, cpDNA) history
Cladistics -- monophyletic clades united by synapomorphies
Parsimony & maximum likelihood approaches
             bootstrapping to assess the robustness of clades (how well-supported are the nodes?)
Molecular clock
             linear extrapolations to estimate divergence times

II. Speciation

Allopatric vs. sympatric vs. parapatric models
Islands as natural laboratories
Biological vs. phylogenetic species concepts
Punctuated equilibrium (macroevolution) vs. gradualism (microevolution)
Role of constraints vs. adaptations in evolution
Wright’s shifting balance theorem vs. Fisher’s mass selection view

Hardy-Weinberg principle: p² + 2pq + q²
        Ratio of homozygotes to heterozygotes
        Gene diversity/expected heterozygosity 1-·p_i²

The 5 forces that drive evolution:
Mutation
Drift (= small population size)
Migration
Non-random mating
Selection (neutral markers/neutral theory)
3 additional assumptions
(Non-overlapping generations)
(Diploid)
(Sexual reproduction)
Measures of genetic structure:
Variance-based (F-statistics)
F_IS (inbreeding) (H_E - H_O)/ H_E = (H_S - H_I)/ H_S
F_ST = (H_T - H_S)/ H_T ;  H_exp at pop. level vs. global 2p-bar*q-bar
Weir and Cockerham�s Q
(R_ST for stepwise mutation model - SMM)
Many ways to derive F-statistics (PHSC, Wahlund, etc.)
Wahlund effect -- excess homozygosity overall whenever gene frequencies
differ across subpopulations
Converse Wahlund effect is "isolate breaking" - excess heterozygosity when divergent
populations interbreed
Genetic distances:
Geometric (no biology)
Cavalli-Sforza chord distance, Rogers� distance
Model-based (assumptions about mutation, drift or migration)
Nei�s (1978), measures based on stepwise mutation model (SMM; e.g., dm²), Reynolds' distance (drift)
Concept of effective population size (N_e) and factors that affect it

Fluctuating population size (harmonic mean)
Sex ratio N_e (4*N_m*N_f)/( N_m + N_f) [4 males + 100 females leads to N_e of 15.4]
Variance in offspring number (> or < Poisson?)
Spatial dispersion (neighborhood size = 4ps2d)
Overlapping generations can reduce N_e

IV. Conservation genetics

Where and when are genetics an appropriate tool?
Rare not equal to depauperate and vice versa
Genetic detective work (e.g., mapping migrants from wintering to breeding grounds)
Some interesting and appropriate applications:
        Delineating worthy taxa
        Assessing information content and biodiversity hotspots.
Neglected importance of quantitative genetics as an evolutionary and conservation tool