We want to reconstruct a history of descent (of organisms or of genes)
To do so, we must be able to identify and study patterns of similarity that are the result of descent from a common ancestor.
Homology - the fundamental concept underlying phylogenetic analysis
Term precedes Darwin, but was based on concept of a bauplan
Homologous parts were equivalent parts within the bauplan.
Evolutionary theory provided a context for the observed patterns of similarity.
Homologous structures are now defined as those that are derived from the same structure in a common ancestor.
Because we are generally uncertain of evolutionary history and wish to reconstruct it from the characteristics of living organisms, it is necessary to look for shared, homologous features. These provide evidence of common ancestry. However, it can at times be difficult to be certain whether or not to structures are homologous.
Remane's criteria
Remane, 1971 (expertly reviewed by Riedl, 1978) provided a theoretical context to help make inferences about homology in cases where the situaion is uncertain.
Three criteria can be used to decide difficult cases (Reidl also gives three additional correlates).
- Positional criterion
- Similar position in comparable systems of features
- Plant examples: axillary tubers, Nepenthes pitchers
- Molecular examples: position of a gene within an operon, location of nucleotide within a sequence.
- Structural criterion
- The structures in question share special features
- Plant examples: adventitious roots, epiphyllous buds
- Molecular examples: sequence similarity of two genes, chemical similarity of amino acids (leucine, isoleucine are often interchanged)
- Transitional criterion
- Intermediate forms can be identified (either phylogenetically or ontogenetically)
- Plant examples: cactus spines (Pereskia), Acacia leaves
- Molecular examples: highly divergent genes (Coleochaete tufA)
When comparing organisms, you can find nested sets of homologies
Homology assessment is incredibly important!
Do not confuse homology with similiarity! Homologous structures can be quite dissimilar, and analogous structures can be quite similar.
To carry information, one must be able to identify homologous characters that show variation
Each character can take two or more character states.
Given a set of homologous characters with different character states, one would like to
Different philosophies in how to treat the data
Overall similarity/dissimilarity ("Phenetic" or distance methods)
We will generally think in terms of dissisimilarity
The most common measure we will use will be the number of substitutions per site
Shared, derived characters ("Cladistic")
Model based methods - use a model of character evolution
Every method has its own set of assumptions
Different approaches can give different answers
Example, given a data matrix with:
three taxa (OTUs)
three characters, each with three character states (0, 1)
The data matrix is composed of homologous characters from different taxa coded for their character state:
t | u | v | w | x | y | z | |
A | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
B | 0 | 0 | 0 | 0 | 1 | 1 | 0 |
C | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
D | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
Three possible rooted trees (plus trichotomy, hybridization, etc).
Phenetic approach: overall dissimilarity
divergence AC=75%, AB=50%, BC=25%
this favors tree I
Cladistic approach: shared derived characters
how can you tell what the derived character state is?
find an outgroup and look at its character states (0,0,0,1)
determine how many changes are necessary to explain trees
trees I and III requre 4 changes, tree II only 3.
this favors tree II
How to prepare a molecular alignment?
Completely objective methods would be best
Such approaches to alignment and phylogenetic analysis are under development, but are not yet fully mature
The best advice at present is to use the most objective method available to create the alignment, then check the results by eye and be sure they are reasonable.
Adjust as necessary, and report what you have done.
Bauplan - literally "plan of construction," (in English we would say blueprint) - refers to the common body organization of a group of organisms. Prior to the development of evolutionary theory, the similarity among organisms that we now understand to be the result of descent from a common ancestor was interpreted as variation around a common structural plan. Intragroup variation was interpreted as deviation from the perfect form for that group. A modern implementation of the concept of bauplan can be seen in the AKC Standards for different breeds of dogs. A dog is considered to be of high quality if it conforms closely to the standard for that breed.
Homologous - Two parts are homologous if they are derived from the same part in a common ancestor.
Analogous - Two parts are analogous if they are similar, but are not derived from the same part in a common ancestor.
Character - A feature that can be determined to be homologous in two or more organisms.
Character state - A specific case or condition of a character.
A.M. Dean. 1998. The molecular anatomy of an ancient adaptive event. Amer. Sci. 86:26-37.
Reeck, G. R., et al. 1987. "Homology" in proteins and nucleic acids: a terminological muddle and a way out of it. Cell 50:667.
Riedl, Rupert 1978. Order in Living Organisms: a Systems Analysis of Evolution. Wiley, Chichester, NY.
Remane, A. 1971. Die Grundlagen des Naturlichen Systems der Vergleichenen Anaotomie udn der Phylogenetik (2nd ed.) Koeltz, Konigstein-Jaunus.
CBMG 688N/O Home | Syllabus | Links | Hints |