“Evolutionary speculation constitutes a kind of metascience, which has the same intellectual fascination for some biologists that metaphysical speculation.

“Evolutionary speculation constitutes a kind of metascience, which has the same intellectual fascination for some biologists that metaphysical speculation possessed for some mediaeval scholastics. It can be considered a relatively harmless habit, like eating peanuts, unless it assumes the form of an obsession; then it becomes a vice” (Stanier, 1970)

Linnaean classification• Two major characteristics

• Kingdom Animalia• Phylum Chordata• Class Mammalia• Order Primates• Family Hominidae• Genus Homo• Species sapiens

BIOL E-127 – 10/01/07

Tree of Life: primary divisions

Tree of Life: primary divisions

Tree of Life: three “domains”• Based on 16S rRNA (Woese, 1987):

Tree of Life: three “domains”• Based on 16S rRNA (Pace, 1997):

Tree basics: meaning

Tree basics: rotation

Tree basics: shape

Tree basics: lengths, unrooted

cladograms vs. phylograms

Tree basics: character change

Key phylogenetic terms

Key phylogenetic terms

Phenetics vs. cladistics

Lysozyme amino acid changes in unrelated ruminants

Phenetics vs. cladistics

Maximum Parsimony• Parsimony – shortest tree (fewest homoplasies)

Microbial systematics

• Formerly Pseudomonas (partial list): Ralstonia, Burkholderia, Hydrogenophaga, Sphingomonas, Methylobacterium, Cellvibrio, Xanthomonas, Acidovorax, Hydrogenophillus, Brevundimonas, Pandoraea

multi-C C1(& all use methanol)

Molecular phylogenetics• Zuckerkandl & Pauling. 1965. Molecules as documents of

evolutionary history. J Theor Biol. 8:357-366.

• Neutral theory (Motoo Kimura, 1968)

16S rRNA as phylogenetic marker• Why a good molecule?

Process to analyze sequence data

Ortholog vs. paralog?

Good Dataset

[A1, A2, A3, A4] [A1, B2, A3, A4]

Bad Dataset

A B

species 1 species 2 species 3 species 4

A1B1

A2B2 A4

B4A3

B3

1. Collect Sequence DataOrtholog vs. paralog?

2. Sequence Alignment

CGGATAAACCGGATAGACCGCTGATAAACCGGATAC

taxa1taxa2taxa3taxa4

Alignment

3. Choose Models

Ancestral Sequences

Observed Sequences

?Model

Choose “model”

Example: Neighbor Joining (NJ)

4. Choose Methods

Taxa CharactersSpecies A ATGGCTATTCTTATAGTACGSpecies B ATCGCTAGTCTTATATTACASpecies C TTCACTAGACCTGTGGTCCASpecies D TTGACCAGACCTGTGGTCCGSpecies E TTGACCAGTTCTCTAGTTCG

A

B

C

DE

Choose methods: distance-based

A B C D E Species A ---- 4 10 9 8Species B ---- 8 11 10Species C ---- 3 8Species D ---- 5Species E ----

A B C D E Species A ---- 4 10 9 8Species B -19.3 ---- 8 11 10Species C -10 -14.7 ---- 3 8Species D -10.7 -11.3 -16 ---- 5Species E -12.7 -13.3 -12 -14.7 ----

M(AB)=d(AB) -[(r(A) + r(B)]/(N-2)

4. Choose Methods

Maximum Parsimony (MP): Model: Evolution goes through the least number of changes

Maximum Likelihood (ML): L (data| model)

Bayesian Inference

Pr(data)

Pr(model)model)|Pr(datadata)|Pr(model

Markov chain Monte Carlo (MCMC) method for sampling from posterior probability distribution

Discrete character methods

5. Assess Reliability

I. Bootstrap

Re-sampling to produce pseudo-dataset (random weighting)

II. Jacknife

Sampling with replacement

III. Permutation test

Random deletion of sub-dataset

Randomize dataset to build null likelihood distribution

CGATCGTTA

CAATGATAG

CGCTGATAA

CGCTGATCG

taxa1

taxa2

taxa3

taxa4

123456789

Dataset1: 729338554Dataset2: 631981282

…

Dataset1: 1-3-56789Dataset2: 12-45678-

…

10073

Assess reliability

5. Assess ReliabilityExample analysis: ancestry of HIV-1

(Gao et al., 1999)

5. Assess ReliabilityFurther analysis: timing of HIV-1

(Korber et al., 2000. Nature 288:1789-1796)