BIO155: Estimating Evolutionary Tree
-
Phylogeny or phylogenic tree is a graphical
summary of this history
-
Synapomorphy is a homologous trait that is
shared among certain species and is similar because it was modified in a common
ancestor (are shared, derived traits)
o
A type of homology
o
Group all of living organisms into a single
lineage that descended from the same common ancestor
-
Any group that includes an ancestor and all of
its descendants is called a monophyletic group (or clade or lineage)
o
Genetic code helps identify that bacteria and
mammals as members of the same monophyletic group, but it does not help us
distinguish bacteria from eukaryotes
o
Bacteria are identified by synapomorphies such
as cell walls that contain a compound called peptidoglycan; eukaryotes have
synapomorphies such as nuclear envelop
o
All synapomorphies are homologous traits, but
not all homologous traits are synapomorphies.
o
Synapomorphies can be identified at whatever
taxonomic level
§
Populations, species, genera, and phyla
o
To understand:
§
Synapomorphies identify evolutionary branch
point
§
Synapomorphies are nested, meaning that you
moved through time and trace a tree from its root to its tips, each branching
event adds one or more shared, derived traits (hierarchy of branching events)
o
Phylogeny inference methods that use these
principles are called cladistics methods (First determine what is ancient, and
what is derived)
§
Outgroup analysis – the character state in the
group of interest (the ingroup) is compared to the state in a very close relative
that clearly branched off earlier (the outgroup) => confirm with the fossil
record
§
A phylogenetic tree inferred by clustering
synapomophies in this way is called cladogram
-
Morphological similarities like the eyes and
skulls evolve independently in different lineages due to convergent evolution,
which occurs when natural selection favors similar structures as solutions to
problems posed by similar environment
-
Reversals are common in DNA data because there
are only four possible states for each base in a sequence. Other things being
equal, there is a 25% chance that a reversal to the previous state will occur
each time a change occurs at a particular site in DNA
-
The most efficient way to distinguish homology
from homoplasy is to analyze many traits in reconstructing evolutionary
relationships instead of just one or a few
-
Under parsimony, biologists consider the
hypothesis of convergence as the most likely. (Two changes occur, instead of 6
in the homologous theory)
o
Octopuses and vertebrates each have a layer of
light sensitive cells in their eyes; but in octopuses these cells are oriented
toward the opening where light enters the eye, while in vertebrates the light
sensitive cells are oriented in the opposite direction (Octopus moves the eyes
back and forth, change the shape of eyeball to focus, but vertebrate depends in
the muscle that change the shape of the lens
o
When parsimony is applied to phylogeny
inference, the preferred tree is the one that has the least total amount of
evolutionary change (least subject to homoplasy, and more reliable source of
synapomorphies)
-
The phylogeny of Whales
o
Lack of posterior limbs
o
Found a fossil of relative of whales (53.5 myr
ago). These whales had hind limbs and reassemble and extinct group of amphibian
mammals called mesonychains
o
Choosing character: Morphology and Molecules
§
First, choose character to use as data
·
Skeletal features and other morphological
characteristics
·
DNA sequence and other molecular traits
o
Finding the best tree implied by the data
§
Parsimony with a single morphological character
§
Cetaceans are close relative to ungulates
·
Ungulates – Perissodactyls (horses/ rhinos),
artiodactyls (cow, deer, pig)
·
Features found in angle bone called astragalus
(in artiodactyl – both end of the bone are smooth and pulley shaped
o
Shows that hippos and whales are not related
o
Whales are descended from artiodactyl ancestor
§
Parsimony with multiple molecular character
·
Milk protein gene called beta-casein
·
Site 142, most organisms seem to have G at this
position from common ancestor
·
Site 192, all taxa have C except the camel
·
Site 162, cows, whales, deer and hippo have a T
at this site (other artiodactyl and outgroup has C)
·
Site 166 provides only synapomorphy in these 60
bases for monophyletic group consisting of whales and hippos
·
Site 177 provides a synapomorphy for a clade
consisting of whales, hippos, pigs, and peccaries, which conflict with info
from site 162
o
Reversal and convergence have resulted in
homoplasy (either 162 or 177 does not reflect the actual evolutionary history
of artiodactyl)
o
Artiodactyl implies total of 47 nucleotide
changes while whale-hippo implies 41 changes.
§
In this case, Whale-Hippo is more parsimonious
§
60 characters and 8 taxa
·
When the number of taxa in a study is relatively
low—fewer than 11—a computer program can evaluate all of the possible trees
(Exhaustive Search) Because it guarantees that the optimal tree implied by a
particular data set will be found, the approach is called an exact method
§
Evaluating the best tree
·
Use other methods besides Parsimony: Maximum
likelihood (ML) and Bayesian Markov Chain Monte Carlo (BMCMC)
o
Maximum likelihood (ML) shows how likely it will
match the hypothesis (deciding how good the data is)
o
Bayesian approaches are similar to likelihood
methods, except that the principle is to ask what the probability is of
particular tree being correct, given the data and a model of how the traits in
question change over time
o
Later on is the introduction of distance methods
o
Produce consensus tree under parsimony
o
Using statistical test o evaluate the best tree
under ML and BMCMC
o
Compare the best tree under parsimony, ML and
BMCMC to see how consistence they are
·
Evaluating Particular Branch: Bootstrapping
o
Not every individual can represent the entire
population (If variety in population is high then less confident that sample
average is reliable)
o
With trees based on ML or BMCMC analyses, you
can compare tree with and without a particular branch and compute which is more
likely.
§
Bootstrapping process begins by randomly
selecting one of the sites and using it as the first entry in a new data set.
Then it randomly selects another site which becomes the second data point in
the new data set.
§
Resampling until particular branches occur in
50%, 80% or 100% of the tree estimated from the resampled data sets
§
Aunder 50%, then polytomy, or a point of
uncertainty
§
Strong support for whale-hippo clade
·
Resolving conflicts
o
Need larger data set and less homoplasy
o
The presence or absence of SINE or LINE at a
homologous location in the genome of two different species can be used as a
trait in phylogeny inference
o
Less likely that SINE will insert itself into
two independent host lineages at exactly the same location (possible but not
improbable)
o
Not likely a reversal because the loss of SINE
or LINE is tracable
o
No homoplasy in data set, and thus no conflicts
among characters when they are mapped onto the tree
o
Found fossils of semi-aquatic ancestor that lead
back to the link between whale and hippo
-
Using Phylogenies to Answer questions
o
Classification and Nomenclature: Is there such a
thing as fish?
§
The effort to name and classify species is
called systematics
§
Linnaean system starts with giving each species
a unique genus and species name and then groups progressively more similar
species into kingdoms, classes, orders, families, and genera (phonetic
approach)
§
In contrast, naming schemes that are based on
evolutionary relationships are referred to as phylogenetic or cladistics
approaches (tree based and branching events).
·
More specifically, only monophyletic groups,
which includes all descendants of common ancestors are named
·
Paraphyletic groups, which include some but not
all descendants of a common ancestor, should not be named
o
Using molecular clock ( a hypothesis)
§
There are good theoretical reasons to expect
that at least some types of DNA sequences change in a clocklike fashion
§
Instead of being favor or eliminated by natural
selection, these “neutral change” responds to a random process called genetic
drift
§
Neutral theory of molecular evolution predicts
that neutral changes in DNA should accumulate in population as rate equal to
mutation rate. If the mutation rate does not change much over time and if
generation times remain similar, then the number of neutral molecular
differences between two taxa should be proportional to the age of their most
recent common ancestors
§
By documenting the number of different neutral
mutations observed in two species and multiplying by a calibration rate,
representing how frequently neutral changes occur per myrs, researchers can
estimate when the species diverged
§
Mutation rate to neutral alleles will vary from
gene to gene, and linkage to linkage and base to base
§
Third position of codons are much more neutral
than the first and second, more like a clocklike rate
§
If alleles change through natural selection,
then the mutations are likely to be working in clock fashion
§
Rate of change that are calibrated for a
particular gene and lineage are unlikely to work for other groups, diff in
generation time and selection histories
§
Determine by fossil and geological record
o
Analyzing Phylogeography: How did Chameleons Get
from Africa to India?
§
The effort to understand where organisms live
and how they came to be there is called biogeography
§
Broke up of Gondwana
§
Human cause
o
Co-speciation: When new species of Aphids Form,
what happens to the Bacteria that live inside their cells?
§
When natural selection occurs during these
interactions and produces adaptations in both species involved, coevolution is
said to take place
§
Aphids and bacteria inside them have symbiotic
relationship (endosymbiosis)
ADVERTISEMENT:
No comments:
Post a Comment