Biological Classification

Origins of classification systems

Polynomial system (before 1750). Binomial system -- Developed by Carl Linnaeus (1750s)

Taxonomic Hierarchy -- Biological systems of classification are hierarchical

Phylogenetics and Systematics.

Systematics: Study of historical evolutionary & genetic relationships among organisms, similarities & differences.

Classification: Ordering of organisms into groups or sets on the basis of their relationships.

Taxonomy: Naming and assignment of organisms to taxa

How do you determine relationships between organisms???

Need to find homologous characters.

Classic example of homology: All tetrapods share homologous 5-digit limb structure even though they perform different functions.
 

Wings of birds and bats are analagous.Structurally different: Bird wing supported by digit #2, bat wing by digit 2-5. Bird wing covered by feathers, bat with skin.
 
 
 
 

The eukaryotic cell is homologous among all species. However, because it is an ancestral character, it does not carry phylogenetic information.
 
 
 
 
 
 
 
 
 
 

Systematics Schools

1. Evolutionary School of Classification-- Based on observed similarities/ differences among groups or organisms evaluated in light of inferred evolutionary history.

Produces phylogram.

Procedures

1. Sorting individuals into phenotypic groups

2. Assigning populations to species; naming species. Compare differences within versus between groups:

3. Grouping species into higher taxa and naming these. Involves:

i. Determination of nearest relatives.

ii. Determine which clusters to recognize as genera, which as species groups.

iii. Arrange genera into groups of progressively higher ranks.


Discovering nearest relative depends on relationship (= similarity to pheneticists; genealogy to cladist; genetic similarity to evolutionary systematist).

4. Delimitation of higher taxa. Considerations:

i. Distinctness of group.

ii Evolutionary role &uniqueness in adaptive zone.

iii. Degree of difference among groups relative to 'dispersion';

iv. Size of taxon;

v. Ranking of related taxa at equivalent level.

Criticisms.

a. Attempts too much -- simultaneously classify, name, determine similarity, map relationships

b. Lacks objectivity -- Relied more on intuition.

c. Uses circular reasoning. Common lineage inferred from resemablance & homology; homology determined by common lineage.

d. Uses unstandardized taxonomic ranks -- genera, families etc in different groups not equivalent.

e. Accepts paraphyletic groups (cladists recognize only monophyletic groups).

f. Uses paraphyletic (not holophyletic) classification (eg classification of crocodiles and birds).

 In conclusion, subjective, arbitrary, probably unstable.

2. Phenetics:

Classification by resemblance--only real thing is what you can measure.

Use operational taxonomic unit (OTUs) -- avoids preconceived ideas.

Greatly facilitated by computerization. Most commonly used measures.

Numerous characters evaluated based on phenetic similarity: Greater the content of information better a given classification will be.

A priori, every character of equal weight in creating natural taxa.

Each character measured numerically..

Phylogenetic inference can be made from taxonomic structures of group using similarity/ clustering programs.

Taxonomy viewed & practiced as empirical science.

Produces phenogram.

 Procedure: 1. Unit characters -- attribute of an organism about which a single statment can be made.

2. Character selection: Characters described by states--continuous or discrete. Can be arranged in transformation series.

3. Weighting--method for determining relative importance of information content of each character.

Analyses

1. Association Coefficients--Coefficients of Similarity

2. Coefficients of Dissimilarity

3. Distance Measures: Methods Based on Pairwise Distances

a) Additive distances.-- work fine if data additive Neighbor-Joining methods. Transforms distance matrix, then uses cluster analysis to infer tree.
 
 
 
 
 
 
 
  b) Ultrametric distances--every common ancestor equidistant from descendants. If it is happening, then can do cluster analysis: Criticisms. a. Generates unnatural and uniformative groups -- phenograms can differ if data analyzed differently. Not every feature is useful in determining relationships.

b. Inadequate character choice and wieghting -- many characters may be convergent.
 
 

 3. Cladistics = Phylogenetic Systematics

Hennig

a) Character is collection of mutually exclusive states which have a fixed order of evolution such that each state is derived directly from just one other state.

b) Monophyletic group: All taxa possess derived state; no taxon outside that group possess that state.

Numerical Cladistics (Parsimony, Compatability, Maximum Likelihood).

Based on Evolutionary homology criterion:

Assumes evolutionarily homologous traits all covary with phylogeny.

-> need method of reconstructing phylogeny that is independent of assumptions of phylogenetic history.

 Taxa could not be grouped based on overall similarity because: a) Similarity in general homologous traits (= plesiomorphies);

b) Similarity due to convergent or parallel evolution (homoplasy).

c) similarity in special homologous traits (= apomorphies).

SO must distinguish:

a) special from general traits;

b) homology from homoplasy

Procedure: Solution provided by Hennig:

General homology can be distinguished from special homology by applying "outgroup criterion": General trait--any trait in 31 member of study group AND in species outside study group;

Special traits--only within study group--grouped by those traits they share.

All evolutionary homologies covary, while homoplasies do not, so:

pattern of relationship supported by largest subset of special similarities is phylogenetic relationship.

 So, for phylogenetic systematics: 1. Assume homology, a priori

2. Use outgroup comparisons to distinguish special from general homologous traits;
 
 
 

Outgroups. In (a) species 1-4 have the character states as given. We wish to know if a or a' was the state in their  common ancestor. In (b) we look at a closely related species, the outgroup. It has a state a, and so we infer that was the state in the ancestor of species 1-4. Dotted line indicates branching relations uncertain. 
 

 
3. group according to shared special homologous traits (synapomorphies).

4. in event of conflicting evidence, choose phylogenetic relationships supported by largest number of traits.

5. Interpret inconsistent results, post hoc, as homoplasies.

So homologies (which indicate phylogenetic relationship) are determined without reference to phylogeny, while homoplasies are determined by reference to phylogeny.

 Comparative Method. Reconstructing phylogenetic relationships between species, then using relationships as a template for explanations of evolutionary origin.

Circularity: If homologies are determined by phylogeny, how can they be used to determine phylogeny? Hennig got round this by assuming:

1) homologous characters outnumber others.

2) homologies determined without reference to phylogeny; homoplasies determined as such BY reference to phylogeny.

3) Each character independently assigned particular homology status depending on properties of species for which the phylogenetic relationships are not being assessed (outgroups).
 
 

 Analyses:

Maximum parsimony. Parsimony methods search for minimum-length trees.

If character polarities could always be reliably assessed, and if no homoplasy in data, then phylogeny reconstruction would amount to nothing more than grouping taxa according to shared derived character states (synapomorphies), with no relevance being attached to the sharing of ancestral states (symplesiomorphies). But usually must invoke parsimony criterion--fewest assumptions of homoplasy.

-> no real difference between computerized methods and manual methods for choosing trees under parsimony criterion.

--Infer topology of tree and character polarities simultaneously, rather than going through 2-stage process of (1) assigning polarities, and (2) estimating tree.

Order defines the nature of permitted character-state transformations. Polarity refers to direction of character evolution.

 Criticisms.
a. Arbitrary decisions in terms of word usage.
b. Misleading conseptualization of ranking -- doesn't have separate 'grouping' and 'ranking' steps.

c. Opertaional neglect of evident facts. Neglects how far taxa have deviated from ancestors.
 
 
 
 
 
 
 
 

 Types of Characters

DNA sequences

 
 
 

Phylogenetic relationships of hominoids as revealed by DNA hybridization.
 
 
 
 
 
 
 
 
 

Relationships of Homo, other great apes & Ramapithecus according to (a) paleontological, &(b) molecular evidence. Dotted lines indicate uncertainty.
 
 
 
 
 
 

Vicariance Biogeography.

Vicariance = geographic separation of continuous biota into 3 2 geographic subunits. May give new species or discrete populations.

Biogeography is study of distribution of organisms in space through time.

Vicariance biogeography = phylogenetic splitting events resulting from geographic separation and resulting parallel phylogenies generated from species subject to similar events.

1. Track synthesis -- search for replicated patterns of distribution among different monophyletic taxa.. Generate area cladograms (replace taxa by area they occupy), then determine:

2. Congruence between area cladograms -- if they agree, indicates generality of vicariant event.
 
 

 Evidence for vicariance biogeography:

Biogeography of chironomid midges in southern hemisphere. Midges live in regions indicated in black.

Area cladogram for spiders in Hawaii based on 12S mitochondrial DNA
 


 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Comparison of Methods


 
 
 

Phenetic and phylogenetic trees can differ.

Phenetic, phylogenetic and evolutionary classifications differ in groups recognized and characters used.
 
 
 
 
 
 
 
 
 
 
 
 
 

4. Evolutionary Systematics--Incoprotates information from both phenetics and cladistics.