Phylogeny: Reconstructing Evolutionary Trees

1
PhylogenyReconstructing Evolutionary Trees

• Chapter 14

2
Phylogenetic trees

• The phylogeny of a group of taxa (species, etc.)
  is its evolutionary history
• A phylogenetic tree is a graphical summary of
  this history indicating the sequence in which
  lineages appeared and how the lineages are
  related to one another
• Because we do not have direct knowledge of
  evolutionary history, every phylogenetic tree is
  an hypothesis about relationships
• Of course, some hypotheses are well supported by
  data, others are not

3
Questions

• How do we make phylogenetic trees?
• Cladistic methodology
• Similarity (phenetics)
• What kinds of data do we use?
• Morphology
• Physiology
• Behavior
• Molecules
• How do we decide among competing alternative
  trees?

4
Similarity

• The basic idea of phylogenetic reconstruction is
  simple
• Taxa that are closely related (descended from a
  relatively recent common ancestor) should be more
  similar to each other than taxa that are more
  distantly related so, all we need to do is
  build trees that put similar taxa on nearby
  branches this is the phenetic approach to tree
  building
• Consider, as a trivial example, leopards, lions,
  wolves and coyotes all are mammals, all are
  carnivores, but no one would have any difficulty
  recognizing the basic similarity between leopards
  and lions, on the one hand, and between wolves
  and coyotes, on the other, and producing this
  tree which, it would probably be universally
  agreed, reflects the true relationships of these
  4 taxa

5
Causes of similarity

• Things are seldom as simple as in the preceding
  example
• We need to consider the concept of biological
  similarity, and the way in which similarity
  conveys phylogenetic information, in greater
  depth
• Homology
• Homoplasy

6
Homology

• A character is similar (or present) in two taxa
  because their common ancestor had that character
• In this diagram, wings are homologous characters
  in hawks and doves because both inherited wings
  from their common winged ancestor

7
Homoplasy

• A character is similar (or present) in two taxa
  because of independent evolutionary origin (i.e.,
  the similarity does not derive from common
  ancestry)
• In this diagram, wings are a homoplasy in hawks
  and bats because their common ancestor was an
  un-winged tetrapod reptile. Bird wings and bat
  wings evolved independently.

8
Types of homoplasy

• Convergence
• Independent evolution of similar traits in
  distantly related taxa streamlined shape,
  dorsal fins, etc. in sharks and dolphins
• Parallelism
• Independent evolution of similar traits in
  closely related taxa evolution of blindness in
  different cave populations of the same fish
  species
• Reversal
• A character in one taxon reverts to an earlier
  state (not present in its immediate ancestor)

9
Reversal

• A character is similar (or present) in two taxa
  because a reversal to an earlier state occurred
  in the lineage leading to one of the taxa
• In this diagram, hawks and cats share the
  ancestral nucleotide sequence ACCT, but this is
  due to a reversal on the lineage leading to cats

hawk
bat
cat
ACCT
ACTT
ACCT
10
Cladistics

• By definition, homology indicates evolutionary
  relationship when we see a shared homologous
  character in two species, we know that they share
  a common ancestor
• Build phylogenetic trees by analyzing shared
  homologous characters
• Of course, we still have the problem of deciding
  which shared similarities are homologies and
  which are homoplasies (to which we shall return)

11
Two kinds of homology 1

• Shared ancestral homology a trait found in all
  members of a group for which we are making a
  phylogenetic tree (and which was present in their
  common ancestor) symplesiomorphy
• For example a backbone is a shared ancestral
  homology for dogs, humans, and lizards
• Symplesiomorphies DO NOT provide phylogenetic
  information about relationships within the group
  being studied

12
Two kinds of homology 2

• Shared derived homology a trait found in some
  members of a group for which we are making a
  phylogenetic tree (and which was NOT present in
  the common ancestor of the entire group)
  synapomorphy
• For example hair is (potentially) a shared
  derived homology in the group dogs, humans,
  lizards
• Synapomorphies DO provide phylogenetic
  information about relationships within the group
  being studied
• In this particular case, if hair is a
  synapomorphy in dogs and humans, then dogs and
  humans share a common ancestor that is not shared
  with lizards, and the common dog-human ancestor
  must have lived more recently than the common
  ancestor of all three taxa

13
A tree for dogs, humans, lizards 1

• The TWO major assumptions that we are making when
  we build this tree are
• hair is homologous in humans and dogs
• hair is a derived trait within tetrapods

14
A tree for dogs, humans, lizards 2

• In the absence of other information, the
  assumption of homology of hair in humans and dogs
  is justified by parsimony (fewest number of
  evolutionary steps is most likely simplest
  explanation)
• Also we can check to see that hair is formed in
  the same way by the same kinds of cells, etc.

15
A tree for dogs, humans, lizards 3

• These trees (in which hair is considered a
  homoplasy in dogs and humans) are less
  parsimonious than the one on the previous slide,
  because they require two independent evolutionary
  origins of hair

16
Character Polarity

• Whats the basis for our second major assumption
  that hair is a derived trait within this group
  (and that absence of hair is primitive)?
• Fossil record
• Outgroup analysis

17
Outgroups 1

• An outgroup is a taxon that is related to, but
  not part of the set of taxa for which we are
  constructing the tree (the in group)
• Selection of an outgroup requires that we already
  have a phylogenetic hypothesis
• A character state that is present in both the
  outgroup and the in group is taken to be
  primitive by the principle of parsimony (present
  in the common ancestor of both the outgroup and
  the in group and, therefore, homologous)

18
Outgroups 2

• In the present example, dog, human, lizard are
  all amniote tetrapods. The anamniote tetrapods
  (amphibia) make a reasonable outgroup for this
  problem
• No amphibia have hair, therefore absence of hair
  amphibia, lizards is primitive (plesiomorphic)
  and presence of hair dogs, humans is derived
  (apomorphic)
• So, presence of hair is a shared derived
  character (synapomorphy), and dogs and humans are
  more closely related to each other than either is
  to lizards

19
A tree for dogs, humans, lizards 4

• The presence of hair is apomorphic (derived)
  because no amphibians have hair

20
Cladistic methodology

• Determine character state polarity by reference
  to outgroup or fossil record
• Construct all possible trees for the taxa in the
  in group
• Map evolutionary transitions in character states
  onto each tree
• Find the most parsimonious tree the one with
  the fewest evolutionary changes
• Only synapomorphies are informative

21
A tree for dogs, humans, lizards 5
(a)
(c)
(b)

• Tree (a) is most parsimonious, so well take that
  as our best estimate of the true phylogeny of
  dog, human, lizard
• Of course, if we studied different characters, or
  used a different outgroup, our phylogenetic tree
  could change

22
The phylogeny of whales

• Based on skeletal characteristics, several
  studies have placed whales (Cetaceans) as close
  relatives of ungulates (hoofed mammals)
  Cetaceans are possibly the sister group of the
  even-toed ungulates (Artiodactyla)
  Artiodactyla hypothesis

23
The Artiodactlya hypothesis for the evolutionary
relationships of Cetacea (Fig. 14.4 a)Odd-toed
ungulates (Perissodactyla horses, rhinos) are
the outgroup
24
The whale hippo hypothesis for the evolutionary
relationships of Cetacea (Fig. 14.4 a)This tree
was proposed based on nucleotide sequence of a
milk protein gene
25
Sequence data for parsimony analysis (Fig.
14.6)Blue shaded bars represent invariant
(uninformative sites, but note error for site
192), and red shaded bars represent
synapomorphies (note, site 177 does not agree
with tree as drawn). Tree is based on parsimony
26
Which phylogeny for whales, if either, is correct?

• According to the whale hippo hypothesis, whales
  are artiodactyls not the sister group to
  artiodactyls
• Artiodactyls are defined by a particular
  adaptation of the astragalus, an ankle bone
• Since modern whales dont have legs, they dont
  have ankle bones, so without more data its hard
  to resolve the conflict between these two
  phylogenetic hypotheses

27
Whale phylogeny more molecular data(Nikaido et
al. 1999)

• SINEs and LINEs Short Interspersed Elements and
  Long Interspersed Elements
• Transposable elements present in hundreds of
  thousands of copies in mammalian genomes
  transposition is relatively infrequent
• Independent transposition into the same location
  in two different genomes is unlikely (homoplasy)
• Therefore, if SINEs and LINEs are present at the
  same location in two taxa, it is most likely
  homologous.

28
Presence/absence of SINEs and LINEs at 20 loci in
a whale (Bairds beaked whale) and six
artiodactyls(Nikaido et al. 1999) (Fig. 14.8)
29
Presence/absence of SINEs and LINEs at 20 loci in
a whale (Bairds beaked whale) and six
artiodactyls(Nikaido et al. 1999) (Fig. 14.8)
30
Presence/absence of SINEs and LINEs at 20 loci in
a whale (Bairds beaked whale) and six
artiodactyls(Nikaido et al. 1999) (Fig. 14.8)
31
Presence/absence of SINEs and LINEs at 20 loci in
a whale (Bairds beaked whale) and six
artiodactyls(Nikaido et al. 1999) (Fig. 14.8)
32
Presence/absence of SINEs and LINEs at 20 loci in
a whale (Bairds beaked whale) and six
artiodactyls(Nikaido et al. 1999) (Fig. 14.8)
33
Whale phylogeny more fossilsIchthyolestes,
Pakicetus, Ambulocetus, Rhodocetus whale-like
ear bones artiodactyl-like astragalusWhales
are an evolutionary line of artiodactyls The
whale hippo tree is supported by additional data