Reconstructing and Using Phylogenies 16 - W.H. Freeman

Reconstructing andUsing Phylogenies 16W.H. Freeman & Company and Sinauer Associates, Inc ©2010Green fluorescent protein (GFP) wasdiscovered in 1962 when Osamu Shimomura,an organic chemist and marinebiologist, led a team that was ableto purify the protein from the tissues ofthe bioluminescent jellyfish Aequoreavictoria. Some 30 years after its initialdiscovery, Martin Chalfie had the idea(and the technology) to link the genefor GFP to other protein-coding genes,so that the expression of specific genesof interest could be visualized in glowinggreen within cells and tissues ofliving organisms (see Figure 13.6). Thiswork was extended by Roger Tsien,who changed some of the amino acidswithin GFP to create proteins of severaldistinct colors. Different coloredproteins meant that the expression ofa number of different proteins couldbe visualized and studied in the sameorganism at the same time. These threescientists were awarded the 2008 NobelPrize in Chemistry for the isolationand development of GFP for visualizinggene expression.Although Tsien was able to producedifferent colored proteins, he could notproduce a red protein. This was frustrating;a red fluorescent protein would beparticularly useful to biologists becausered light penetrates tissues more easilythan do other colors. Tsien’s work stimulatedMikhail Matz to look for new fluorescentproteins in corals (which are relativesof the jellyfishes). Among the differentspecies he studied, Matz found coral proteinsthat fluoresced in various shades ofgreen, cyan (blue-green)—and red.How had fluorescent red pigmentsevolved among the corals, given thatthe necessary molecular changes hadeluded Tsien? To answer this question,Matz sequenced the genes ofthe fluorescent proteins and usedthese sequences to reconstruct theevolutionary history of the amino acidchanges that produced different colors indifferent species of corals.Matz’s work showed that the ancestralfluorescent protein in corals was green,and that red fluorescent proteins evolvedin a series of gradual steps. His analysis allowedhim to retrace these steps in detail.Such an evolutionary history, as depictedin a tree of relationships among lineages,is called a phylogeny.The evolution of many aspects of anorganism’s biology can be reconstructedusing phylogenetic methods. This informationis used in all fields of biology tounderstand the structure, function, andbehavior of organisms.QHowQUESTIONare phylogenetic methodsused to resurrect proteinsequences from extinct organisms?The reef-building coralAcropora millepora showscyan and red fluorescence.This photograph was takenunder a fluorescent microscopethat affects the colorswe see; the colors are perceiveddifferently by oceananimals, depending in part onthe depth of the water.kEY CONCEPTS16.1 All of Life Is Connectedthrough Its EvolutionaryHistory16.2 Phylogeny Can Be Reconstructedfrom Traits ofOrganisms16.3 Phylogeny MakesBiology Comparativeand Predictive16.4 Phylogeny Is the Basis ofBiological Classification

310 Chapter 16 | Reconstructing and Using PhylogeniesW.H. Freeman & Company and Sinauer Associates, Inc ©2010concept16.1The sequencing of complete genomes from many diverse specieshas confirmed what biologists have long suspected: all of life isrelated through a common ancestor. The common ancestry of lifeexplains why the general principles of biology apply to all organisms.Thus we can learn much about how the human genomeworks by studying the biology of model organisms because weshare a common evolutionary history with those organisms. Theevolutionary history of these relationships is known as phylogeny,and a phylogenetic tree is a diagrammatic reconstructionof that history.Phylogenetic trees are commonly used to depict the evolutionaryhistory of species, populations, and genes. For manyyears such trees have been constructed based on physical structures,behaviors, and biochemical attributes. Now, as genomesare sequenced for more and more organisms, biologists are ableto reconstruct the history of life in ever greater detail.In Chapter 15, we discussed why we expect populations oforganisms to evolve over time. Such a series of ancestor and descendantpopulations forms a lineage, which we can depict as aline drawn on a time axis:TimeA species, population, orgene at one point in time……becomes a lineage as we followits descendants through time.What happens when a single lineages divides into two? For example,a geographic barrier may divide an ancestral populationinto two descendant populations that no longer interact withone another. We depict such an event as a split, or node, in aphylogenetic tree. Each of the descendant populations give riseto a new lineage, and as these independent lineages evolve, newtraits arise in each:NodeAll of Life Is Connected through ItsEvolutionary HistoryA split occurs when theancestral lineage divides intotwo descendant lineages…As the lineages continue to split over time, this history can berepresented in the form of a branching tree that can be used totrace the evolutionary relationships from the ancient commonancestor of a group of species, through the various lineage splits,up to the present populations of the organisms:TimePOL HillisSinauer AssociatesMorales StudioFigure INTXT16.03 Date 06-17-10PresentA phylogenetic tree may portray the evolutionary history ofall life forms; of a major evolutionary group (such as the insects);of a small group of closely related species; or in some cases, eventhe history of individuals, populations, or genes within a species.The common ancestor of all the organisms in the tree forms theroot of the tree. The depictions of phylogenetic trees in this bookare rooted at the left, with time flowing from left (earliest) toright (most recent):Commonancestor(root)15The positions of the nodes onthe time scale indicate the timesof divergence events.ChimpanzeeHumanGorillaOrangutan10 5 0Time (millions of years ago) PresentTimePOL HillisSinauer AssociatesMorales StudioFigure INTXT16.01 Date 06-17-10…and each lineagecontinues to evolveindependently asdifferent traits(represented by reddots) arise.The timing of splitting events in lineages is shown by the positionof nodes on a time axis. These splits represent events whereone lineage diverged into two, such as a speciation event (for atree of species), a gene duplication event (for a tree of genes), or atransmission event (for a tree of viral lineages transmitted througha host population). The time axis may have an explicit scale, or itmay simply show the relative timing of divergence events.In this book’s illustrations, the order in which nodes are placedalong the horizontal (time) axis has meaning, but the vertical

5.2 16.1 All of Life Is Connected through Its Evolutionary History 311distance between the branches does not. Vertical distances havebeen adjusted for legibility and clarity of presentation; they donot correlate with the degree of similarity or difference betweengroups. Note too that lineages can be rotated around nodes in thetree, so the vertical order of lineages is also largely arbitrary:CommonancestorLampreyPerchW.H. Freeman & Company and Sinauer Associates, Inc ©2010ChimpanzeeHumanGorillaOrangutanHumanChimpanzeeGorillaOrangutanAny group of species that we designate with a name is ataxon (plural taxa). Examples of familiar taxa include humans,primates, mammals, and vertebrates; in this series, each taxon isalso a member of the next, more inclusive taxon. Any taxon thatconsists of all the evolutionary descendants of a common ancestoris called a clade. Clades can be identified by picking any pointon a phylogenetic tree and from that point tracing all the descendantlineages to the tips of the terminal branches (Figure 16.1).Two species that are each other’s closest relatives are called sisterspecies; similarly, any two clades that are each other’s closestrelatives are sister clades.Before the 1980s, phylogenetic trees tended to be seen onlyin the literature on evolutionary biology, especially in the areaof systematics—the study and classification of biodiversity. Butalmost every journal in the life sciences published during the lastfew years contains phylogenetic trees. Trees are widely used inmolecular biology, biomedicine, physiology, behavior, ecology,and virtually all other fields of biology. Why have phylogeneticstudies become so widespread?Phylogenetic trees are the basis ofcomparative biologyPOL HillisIn Sinauer biology, Associates we study life at all levels of organization—fromMorales Studiogenes, Figure INTXT16.05 cells, organisms, Date 06-17-10 populations, and species to the majordivisions of life. In most cases, however, no individual gene ororganism (or other unit of study) is exactly like any other geneor organism that we investigate.Consider the individuals in your biology class. We recognizeeach person as an individual human, but we know that no twoare exactly alike. If we knew everyone’s family tree in detail, thegenetic similarity of any pair of students would be more predictable.We would find that more closely related students havemany more traits in common (from the color of their hair to theirsusceptibility or resistance to diseases). Likewise, biologists usephylogenies to make comparisons and predictions about sharedtraits across genes, populations, and species.The evolutionary relationships among species, as representedin the tree of life, form the basis for biological classification. Biologistsestimate that there are tens of millions of species onEarth. So far, however, only about 1.8 million species have beenclassified—that is, formally described and named. New speciesMammalsAmniotesTetrapodsVertebratesSalamanderLizardCrocodilePigeonMouseChimpanzeeFIGURE 16.1 Clades Represent All the Descendants of aCommon Ancestor All clades are subsets of larger clades, withall of life as the most inclusive taxon. In this example, the groupscalled mammals, amniotes, tetrapods, and vertebrates representsuccessively larger clades. Only a few species within each clade arerepresented on this tree.are being discovered all the time and phylogenetic analyses areconstantly reviewed and revised, so our knowledge of the treeof life is far from complete. Yet knowledge of evolutionary relationshipsis essential for making comparisons in biology, sobiologists build phylogenies for groups of interest as the needarises. The tree of life’s evolutionary framework allows us tomake many predictions about the behavior, ecology, physiology,genetics, and morphology of species that have not yet beenstudied in detail.When biologists compare species, they observe traits that differwithin the group of interest and try to ascertain when thesePrinciples of LIFE SadavaSinauer AssociatesMorales traits Studio evolved. In many cases, investigators are interested in howFigure the 16.01 evolution Date of 06-16-10 a trait relates to environmental conditions or selectivepressures. For instance, scientists have used phylogeneticanalyses to discover changes in the genome of human immunodeficiencyvirus that confer resistance to particular drug treatments.The association of a particular genetic change in HIV witha particular treatment provides a hypothesis about the evolutionof resistance that can be tested experimentally.Any features shared by two or more species that have been inheritedfrom a common ancestor are said to be homologous. Homologousfeatures may be any heritable traits, including DNAsequences, protein structures, anatomical structures, and evensome behavior patterns. For example, all living vertebrates havea vertebral column, as did the ancestral vertebrate. Therefore, thevertebral column is judged to be homologous in all vertebrates.

312 Chapter 16 | Reconstructing and Using PhylogeniesW.H. Freeman & Company and Sinauer Associates, Inc ©2010Derived traits provide evidence of evolutionaryrelationshipsIn tracing the evolution of a character, biologists distinguish betweenancestral and derived traits. Each character of an organismevolves from one condition (the ancestral trait) to another condition(the derived trait). Derived traits that are shared amonga group of organisms and are also viewed as evidence of thecommon ancestry of the group are called synapomorphies (syn,“shared”; apo, “derived”; morph, “form,” referring to the “form”of a trait). Thus the vertebral column is considered a synapomorphy—ashared, derived trait—of the vertebrates. (The ancestraltrait was an undivided supporting rod.)Not all similar traits are evidence of relatedness. Similar traitsin unrelated groups of organisms can develop for either of thefollowing reasons:• Independently evolved traits subjected to similar selectionpressures may become superficially similar, a phenomenoncalled convergent evolution. For example, although the wingbones of bats and birds are homologous, having been inheritedfrom a common tetrapod ancestor, the wings of bats and birdsare not homologous because they evolved independently fromthe forelimbs of different nonflying ancestors (Figure 16.2).• A character may revert from a derived state back to an ancestralstate in an event called an evolutionary reversal. Forexample, the derived limbs of terrestrial tetrapods evolvedfrom the ancestral fins of their aquatic ancestors. Then, withinthe mammals, the ancestors of modern cetaceans (whales anddolphins) returned to the ocean, and cetacean limbs evolved toBat wingBird wingBones shown in the samecolor are homologous.FIGURE 16.2 The Bones Are Homologous, the Wings AreNot The supporting bone structures of both bat wings and birdwings are derived from a common tetrapod (four-limbed) ancestorand are thus homologous. However, the wings themselves—anadaptation for flight—evolved independently in the two groups.once again resemble their ancestral state—fins. The superficialsimilarity of cetacean and fish fins does not suggest a closerelationship between these groups; the similarity arises fromevolutionary reversal.Similar traits generated by convergent evolution and evolutionaryreversals are called homoplastic traits or homoplasies.A particular trait may be ancestral or derived, depending onour point of reference. For example, all birds have feathers. Weinfer from this that feathers (which are highly modified scales)were present in the common ancestor of modern birds. Therefore,we consider the presence of feathers to be an ancestral traitfor any particular group of modern birds, such as the songbirds.However, feathers are not present in any other living animals. Inreconstructing a phylogeny of all living vertebrates, the presenceof feathers is a derived trait found only among birds, and thus isa synapomorphy of the birds.Do You Understand Concept 16.1?• What biological processes are represented in aphylogenetic tree?• Why is it important to consider only homologouscharacters in reconstructing phylogenetic trees?• What are some reasons that similar traits might ariseindependently in species that are only distantlyrelated? Can you think of examples among familiarorganisms?Phylogenetic analyses of evolutionary history have become increasinglyimportant to many types of biological research in recentyears, and they are the basis for the comparative nature ofbiology. For the most part, however, evolutionary history cannotbe observed directly. How, then, do biologists reconstruct thepast?concept16.2Phylogeny Can Be Reconstructedfrom Traits of OrganismsConsider the eight vertebrate animals listed in Table 16.1: lamprey,perch, salamander, lizard, crocodile, pigeon, mouse, andchimpanzee. To illustrate how the phylogenetic tree in Figure16.3 is constructed, we assume initially that a given derived traitarose only once during the evolution of these animals (there hasbeen no convergent evolution), and that no derived traits werelost from any of the descendant groups (there has been no evolutionaryreversal). For simplicity, we have selected traits that areeither present (+) or absent (–).In a phylogenetic study, the group of organisms of primaryinterest is called the ingroup. As a point of reference, an ingroupis compared with an outgroup: a species or group known tobe closely related to but phylogenetically outside the group ofinterest. If the outgroup is known to have diverged before the ingroup,the outgroup can be used to determine which traits of the

5.2 16.2 Phylogeny Can Be Reconstructed from Traits of Organisms 313TABLE 16.1 Eight Vertebrates and the Presence or Absence of Some Shared Derived Traitsderived traitW.H. Freeman & Company and Sinauer Associates, Inc ©2010CLAWS mammary keratinoustaxon jaws lungs or nails gizzard feathers fur glands scalesLamprey (outgroup) – – – – – – – –Perch + – – – – – – –Salamander + + – – – – – –Lizard + + + – – – – +Crocodile + + + + – – – +Pigeon + + + + + – – +Mouse + + + – – + + _Chimpanzee + + + – – + + _ingroup are derived (i.e., evolved within the ingroup) and whichare ancestral (i.e., evolved before the origin of the ingroup). Thelamprey belongs to a group of jawless fishes thought to haveseparated from the lineage leading to the other vertebrates beforethe jaw arose. Therefore, we have specified the lamprey as theoutgroup for our analysis. Because derived traits were acquiredby other members of the vertebrate lineage after they divergedfrom the outgroup, any trait that is present in both the lampreyand the other vertebrates is judged to be ancestral.We begin by noting that the chimpanzee and mouse sharetwo derived traits—mammary glands and fur—that are absentThe earliest branch in the treerepresents the evolutionary splitbetween the outgroup (lamprey)and the ingroup (the remainingspecies of vertebrates).Derived traits areindicated alonglineages in whichthey evolved.CommonancestorJawsLungsClawsor nailsin both the outgroup and in the other species of the ingroup.Therefore, we infer that mammary glands and fur are derivedtraits that evolved in a common ancestor of chimpanzees andmice after that lineage separated from the lineages leading tothe other vertebrates. These characters are synapomorphies thatunite chimpanzees and mice (as well as all other mammals, althoughwe have not included other mammalian species in thisexample). By the same reasoning, we can infer that the othershared derived traits are synapomorphies for the various groupsin which they are expressed. For instance, keratinous scales are asynapomorphy of the lizard, crocodile, and pigeon.FIGURE 16.3 Inferring a Phylogenetic Tree This phylogenetic tree was constructedfrom the information given in Table 16.1 using the parsimony principle. Each cladein the tree is supported by at least one shared derived trait, or synapomorphy.KeratinousscalesGizzardFeathersLamprey(outgroup)PerchSalamanderLizardCrocodilePigeonThe lamprey isdesignated asthe outgroup.IngroupFur; mammaryglandsMouseChimpanzee

314 Chapter 16 | Reconstructing and Using PhylogeniesW.H. Freeman & Company and Sinauer Associates, Inc ©2010Table 16.1 also tells us that, among the animals in our ingroup,the pigeon has a unique trait: the presence of feathers.Feathers are a synapomorphy of birds and their extinct relatives.However, because we only have one bird in this example, thepresence of feathers provides no clues concerning relationshipsamong these eight species of vertebrates. However, gizzards arefound in both birds and crocodiles, so this trait is evidence of aclose relationship between birds and crocodilians.By combining information about the various synapomorphies,we can construct the phylogenetic tree in Figure 16.3. Weinfer from our information that mice and chimpanzees—theonly two animals that share fur and mammary glands—share amore recent common ancestor with each other than they do withpigeons and crocodiles. Otherwise, we would need to assumethat the ancestors of pigeons and crocodiles also had fur andmammary glands but subsequently lost them; these additionalassumptions are unnecessary in this case.This particular tree was easy to construct because it is basedon a very small sample of traits, and the derived traits we examinedevolved only once and were never lost after they appeared.Had we included a snake in the group, our analysis would nothave been as straightforward. We would need to examine additionalcharacters to determine that snakes evolved from a groupof lizards that had limbs. In an evolutionary reversal, limbs werelost in the ancestors of snakes as an adaptation to a subterraneanexistence.Typically, biologists construct phylogenetic trees using hundredsor thousands of traits. With larger data sets, we would expectApply the ConceptPhylogeny can be reconstructed from traits of organismsThe matrix below supplies data for seven land plants and anoutgroup (an aquatic plant known as a stonewort). Each traitis scored as either present (+) or absent (–) in each of the plants.Use this data matrix to reconstruct the phylogeny of land plantsand answer the questions.1. Which two of these taxa are most closely related?to observe traits that have changed more than once, and thuswould expect to see convergence and evolutionary reversal. Howdo we determine which traits are synapomorphies and which arehomoplasies? One way is to invoke the principle of parsimony.yourBioPortal.comGo to WEB ACTIVITY 16.1Constructing a Phylogenetic Treeand INTERACTIVE TUTORIAL 16.1Phylogeny: Evolution of TraitsParsimony provides the simplest explanation forphylogenetic dataIn its most general form, the parsimony principle states that thepreferred explanation of observed data is the simplest explanation.Applying the principle of parsimony to the reconstructionof phylogenies entails minimizing the number of evolutionarychanges that need to be assumed over all characters in all groupsin the tree. In other words, the best hypothesis under the parsimonyprinciple is one that requires the fewest homoplasies. Thisapplication of parsimony is a specific case of a general principleof logic called Occam’s razor: The best explanation is the one thatbest fits the data while making the fewest assumptions. Morecomplicated explanations are accepted only when the evidencerequires them. Phylogenetic trees represent our best estimatesabout evolutionary relationships, given the evidence available.2. Plants that produce seeds are known as seed plants. Whatis the sister group to the seed plants among these taxa?3. Which two traits evolved along the same branch of yourreconstructed phylogeny?4. Are there any homoplasies in your reconstructedphylogeny?PersistentlyProtected green VASCULAR mEgaphyllsTaxON Embryos True roots sporophyte cells Stomata (true leaves) SeedsStonewort (outgroup) – – – – – – –Liverwort + – – – – – –Pine tree + + + + + + +Bracken fern + + + + + + –Club moss + + + + + – –Sphagnum moss + – – – + – –Hornwort + – + – + – –Sunflower + + + + + + +Trait

16.2 5.2 Phylogeny Can Be Reconstructed from Traits of Organisms 315W.H. Freeman & Company and Sinauer Associates, Inc ©2010As with all analyses in science, phylogenetic trees are continuallymodified as additional information is obtained.Phylogenies are reconstructed from manysources of dataNaturalists have constructed various forms of phylogenetic treesfor more than 150 years. In fact, the only figure in the first editionof Darwin’s Origin of Species was a conceptual diagram of a phylogenetictree. Phylogenetic tree construction has been revolutionizedby the advent of computer software that allows us to considerfar more data and analyze far more traits than could ever beforebe processed. Combining these advances in methodology with themassive comparative data sets being generated through genomesequencing and other molecular studies, biologists are learningdetails about the tree of life at a remarkable pace.Any trait that is genetically determined—and therefore heritable—canbe used in a phylogenetic analysis. Evolutionaryrelationships can be revealed through studies of morphology,development, the fossil record, behavioral traits, and moleculartraits such as DNA and protein sequences.morphology An important source of phylogenetic informationis morphology: the presence, size, shape, and other attributesof body parts. Since living organisms have been observed, depicted,and studied for millenia, we have a wealth of recordedmorphological data as well as extensive museum and herbariumcollections of organisms whose traits can be measured. Newtechnological tools, such as the electron microscope and computedtomography (CT) scans, enable systematists to examineand analyze the structures of organisms at much finer scales thanwas formerly possible.Most species are described and known primarily by theirmorphology, and morphology still provides the most comprehensivedata set available for many taxa. The morphologicalfeatures that are important for phylogenetic analysis are oftenspecific to a particular group. For example, the presence, development,shape, and size of various features of the skeletal systemare important in vertebrate phylogeny, whereas the structures ofthe floral organs (petals, carpels, stamens and sepals; see p. 796)are important for studying the relationships among floweringplants.Reconstructing the evolutionary relationships of most extinctspecies depends almost exclusively on morphological comparisons.Fossils show us where and when organisms lived in thepast and give us an idea of what they looked like. Fossils provideimportant evidence that helps distinguish ancestral fromderived traits. The fossil record can also reveal when lineagesdiverged and began their independent evolutionary histories.Furthermore, in groups with few species that have survived tothe present, information on extinct species is often critical to anunderstanding of the large divergences among the survivingspecies. The fossil record is limited, however; for some groups,few or no fossils have been found, and for others the fossil recordis fragmentary.Although useful, morphological approaches to phylogeneticanalysis do have limitations. Some taxa exhibit little morphologicaldiversity, despite great species diversity. For example, the phylogenyof the leopard frogs of North and Central America would bedifficult to infer from morphological differences alone, becausethe many species look very similar, despite important differencesin their behavior and physiology. At the other extreme, few morphologicaltraits can be compared across distantly related species(an earthworm and a mammal, for instance). Furthermore, somemorphological variation has an environmental rather than a geneticbasis and so must be excluded from phylogenetic analyses.For these reasons, an accurate phylogenetic analysis often requiresinformation beyond that supplied by morphology.development Observations of similarities in developmentalpatterns may reveal evolutionary relationships. Some organismsexhibit similarities in early developmental stages only. The larvaeof marine creatures called sea squirts, for example, have a flexiblegelatinous rod in the back—the notochord—that disappears as thelarvae develop into adults. All vertebrate animals also have a notochordat some time during their development (Figure 16.4).This shared structure is one of the reasons for inferring that seasquirts are more closely related to vertebrates than would besuspected if only adult sea squirts were examined.LINK For more on the fascinating role of developmentalprocesses in evolution, see Chapter 14.behavior Some behavioral traits are culturally transmitted(learned from other individuals); others have a genetic basis (seeChapter 41). If a particular behavior is culturally transmitted, itmay not accurately reflect evolutionary relationships (but maynonetheless reflect cultural connections). Bird songs, for instance,are often learned and may be inappropriate traits for phylogeneticanalysis. Frog calls, however, are genetically determinedand appear to be acceptable sources of information for reconstructingphylogenies.molecular data All heritable variation is encoded in DNA,and so the complete genome of an organism contains an enormousset of traits (the individual nucleotide bases of DNA) thatcan be used in phylogenetic analyses. In recent years, DNA sequenceshave become among the most widely used sources ofdata for constructing phylogenetic trees. Comparisons of nucleotidesequences are not limited to the DNA in the cell nucleus.Eukaryotes have genes in their mitochondria as well as in theirnuclei; plant cells have genes in their chloroplasts as well.The chloroplast genome (cpDNA) is used extensively in phylogeneticstudies of plants because it has changed slowly overevolutionary time and can thus be used to study relatively ancientphylogenetic relationships. Most animal mitochondrial DNA(mtDNA) has changed more rapidly, so mitochondrial genes areused to study evolutionary relationships among closely relatedanimal species (the mitochondrial genes of plants evolve moreslowly). Many nuclear gene sequences are also commonly analyzed,and now that entire genomes have been sequenced frommany species, they too are used to construct phylogenetic trees.Information on gene products (such as the amino acid sequencesof proteins) is also widely used for phylogenetic analyses.

316 Chapter 16 | Reconstructing and Using PhylogeniesSea squirt larvaAdultNeural tubeNotochordSea squirt and frog larvae (tadpoles) shareseveral morphological similarities, includingthe presence of a notochord for body support.W.H. Freeman & Company and Sinauer Associates, Inc ©2010Frog larvaNeuraltubeNotochordFIGURE 16.4 The Chordate Connection Embryonic developmentcan offer vital clues to evolutionary relationships, since larvae sometimesshare similarities that are not apparent in the adults. An example is thenotochord, a synapomorphy of the chordates (a taxonomic group thatincludes the sea squirts as well as vertebrates such as frogs). All chordateshave a notochord during their early development. The notochordis lost in adult sea squirts, whereas in adult frogs—as in all vertebrates—the vertebral column replaces the notochord as the body’s supportstructure.Mathematical models expand the power ofphylogenetic reconstructionAs biologists began to use DNA sequences to infer phylogenies,they developed explicit mathematical models describinghow DNA sequences change over time. These models accountfor multiple changes at a given position in a DNA sequence.They also take into account different rates of change at differentpositions in a gene, at different positions in a codon, andamong different nucleotides. For example, transitions (changesbetween two purines or between two pyrimidines) are usuallymore likely than are transversions (changes between a purine andpyrimidine). Principles of LIFE SadavaSinauer AssociatesMathematical Figure 16.04 Date models 06-16-10 can be used to compute how a treemight evolve given the observed data. A maximum likelihoodmethod will identify the tree that most likely produced the observeddata, given the assumed model of evolutionary change.Maximum likelihood methods can be used for any kind of characters,but they are most often used with molecular data, forwhich explicit mathematical models of evolutionary change areeasier to develop. The principal advantages to maximum likelihoodanalyses are that they incorporate more information aboutevolutionary change than do parsimony methods, and they areeasier to treat in a statistical framework. The principal disadvantagesare that they are computationally intensive and requireexplicit models of evolutionary change (which may not be availablefor some kinds of character change).AdultDespite the similarity of theirlarvae, the morphology of adultfrogs and sea squirts provideslittle evidence of the commonancestry of these two groups.The accuracy of phylogenetic methods canbe testedIf phylogenetic trees represent reconstructions of past events,and if many of these events occurred before any humans werearound to witness them, how can we test the accuracy of phylogeneticmethods? Biologists have conducted experiments bothin living organisms and with computer simulations that havedemonstrated the effectiveness and accuracy of phylogeneticmethods.In one experiment designed to test the accuracy of phylogeneticanalysis, a single viral culture of bacteriophage T7 wasused as a starting point, and lineages were allowed to evolvefrom this ancestral virus in the laboratory (Figure 16.5). Theinitial culture was split into two separate lineages, one of whichbecame the ingroup for analysis and the other of which becamethe outgroup for rooting the tree. The lineages in the ingroupwere split in two after every 400 generations, and samples ofthe virus were saved for analysis at each branching point. Thelineages were allowed to evolve until there were eight lineagesin the ingroup. Mutagens were added to the viral cultures toincrease the mutation rate so that the amount of change and thedegree of homoplasy would be typical of the organisms analyzedin average phylogenetic analyses. The investigators thensequenced samples from the end points of the eight lineages, aswell as from the ancestors at the branching points. They thengave the sequences from the end points of the lineages to other

5.2 16.2 Phylogeny Can Be Reconstructed from Traits of Organisms 31716.5 The Accuracy of Phylogenetic Analysis To test whetheranalysis of gene sequences can accurately reconstruct evolutionaryphylogeny, we must have an unambiguously known phylogeny toINVESTIGATIONcompare against the reconstruction. Will the observed phylogeny matchthe reconstruction?HYPOTHESISA phylogeny reconstructed from analysis of the DNA sequences of living organisms canaccurately match the known evolutionary history of the organisms.W.H. Freeman & Company and Sinauer Associates, Inc ©20101 Select single placque(source of commonancestor).2 Split each lineageevery 400 generations,sequencing eachancestor at time of split.METHODIn the laboratory, one group of investigators produced an unambiguous phylogeny of 9 virallineages, enhancing the mutation rate to increase variation among the lineages.GenerationsGrowth inpresenceof mutagen400 400400OutgroupA lineageD lineageC lineageE lineageF lineageH lineageB lineageG lineage3Present final genes (blue dots) to a secondgroup of investigators who are unaware ofthe history of the lineages or the genesequences of the ancestral viruses. These"blind" investigators then determine thesequences of the descendant genes anduse these sequences to reconstruct theevolution of these lineages in the form of aphylogenetic tree.RESULTSThe true phylogeny and ancestral DNA sequences were accurately reconstructed solely from the DNAsequences of the viruses at the tips of the tree.CONCLUSIONPhylogenetic analysis of DNA sequences can accurately reconstruct evolutionary history.ANALYZE THE DATAThe full DNA sequences for the T7 strains in this experiment are thousands of nucleotides long. Thenucleotides (”characters”) at 23 DNA positions are given in the table.Character at position1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23Outgroup C C G G G C C T C C T C G A C C G G C A C G GA T C G G G C C C C C C C A A C C G A T A C A AB C C G G G T C C C T C C G A T T A G C G T G GC C C G G G C C C T C C T A A C C G G T A C A AD T C A G G C C C C C C C A A C C G A T A C A AE C T G G G C C C C C C T A A C C G G T A C A AF C T G A A C C C C C C C G A C T G G C G C G GG C C G G G T T C C T C C G A T T A G C G C G GH C C G G A C C C C C C C G C C T G G C G C G GA. Construct a phylogenetic tree from these nucleotide DNA positions using the parsimony method. Use the outgroup to rootyour tree. Assume that all changes among nucleotides are equally likely.B. Using your tree, reconstruct the DNA sequences of the ancestral lineages.For more, go to Working with Data 16.1 at yourBioPortal.com.Go to yourBioPortal.com for original citations, discussions, and relevant links for all INVESTIGATION figures.POL HillisSinauer AssociatesMorales StudioFigure 16.05 Date 06-17-10

318 Chapter 16 | Reconstructing and Using PhylogeniesW.H. Freeman & Company and Sinauer Associates, Inc ©2010investigators to analyze, without revealing the known history ofthe lineages or the sequences of the ancestral viruses.After the phylogenetic analysis was completed, the investigatorsasked two questions. Did phylogenetic methods reconstructthe known history correctly? And, were the sequences ofthe ancestral viruses reconstructed accurately? The answer inboth cases was “yes.” The branching order of the lineages wasreconstructed exactly as it had occurred; more than 98% of thenucleotide positions of the ancestral viruses were reconstructedcorrectly; and 100% of the amino acid changes in the viral proteinswere reconstructed correctly.yourBioPortal.comGo to ANIMATED TUTORIAL 16.1Using Phylogenetic Analysis to ReconstructEvolutionary HistoryThe experiment shown in Figure 16.5 demonstrated that phylogeneticanalysis was accurate under the conditions tested, butit did not examine all possible conditions. Other experimentalstudies have taken other factors into account, such as the sensitivityof phylogenetic analysis to convergent environments andhighly variable rates of evolutionary change. In addition, computersimulations based on evolutionary models have been usedextensively to study the effectiveness of phylogenetic analysis.These studies have also confirmed the accuracy of phylogeneticmethods and have been used to refine those methods and extendthem to new applications.Do You Understand Concept 16.2?• How is the parsimony principle used in reconstructingevolutionary history?• Why is it important to consider the entire life cyclewhen reconstructing an organism’s evolutionaryhistory?• What are some comparative advantages anddisadvantages of morphological and molecularapproaches for reconstructing phylogenetictrees?• Contrast experimental and simulation approachesfor testing the accuracy of phylogenetic reconstructionsof evolutionary history. Can you think of someaspects of phylogenetic accuracy that might be morepractical to test using computer simulation than withexperimental studies of viruses?Why do biologists expend the time and effort necessary to reconstructphylogenies? In fact, information about the evolutionaryrelationships among organisms is a useful source of datafor scientists investigating a wide variety of biological questions.Next we will describe how phylogenetic trees are used to answerquestions about the past, and to predict and compare traits oforganisms in the present.concept16.3Phylogeny Makes BiologyComparative and PredictiveOnce a phylogeny is reconstructed, what do we do with it? Whatbeyond an understanding of evolutionary history does phylogenyoffer us?Reconstructing the past is important forunderstanding many biological processesPhylogeny often clarifies the origin and evolution of traits thatare of great interest in understanding fundamental biologicalprocesses. This information is then widely applied in fields suchas agriculture and medicine.self-compatibility Like most animals, most flowering plants(angiosperms) reproduce by mating with another individualof the same species. But in many angiosperm species, the sameindividual produces both male and female gametes (containedwithin pollen and ovules, respectively). Outcrossing species havemechanisms to prevent fertilization of the ovule by the individual’sown pollen, and so are referred to as self-incompatible. Individualsof some species, however, regularly fertilize their ovulesusing their own pollen; they are self-fertilizing or selfing species,and their gametes are self-compatible.The evolution of angiosperm fertilization mechanisms was examinedin Leptosiphon (formerly called Linanthus; see page 323),a genus in the phlox family that exhibits a diversity of breedingsystems and pollination mechanisms. The outcrossing (self-incompatible)species of Leptosiphon have long petals and are pollinatedby long-tongued flies. In contrast, self-pollinating specieshave short petals and do not require insect pollinators to reproducesuccessfully. Using nuclear ribosomal DNA sequences, investigatorsreconstructed a phylogeny for twelve species in thegenus (Figure 16.6). They then determined whether each specieswas self-compatible by artificially pollinating flowers withthe plant’s own pollen or with pollen from other individuals andobserving whether viable seeds formed.The reconstructed phylogeny suggests that self-incompatibilityis the ancestral state and that self-compatibility evolvedthree times within this group of Leptosiphon. The change to selfcompatibilityeliminated the plants’ dependence on an outsidepollinator and has been accompanied by the evolution of reducedpetal size. Indeed, the striking morphological similarity ofthe flowers in the self-compatible groups once led to their beingclassified as members of a single species. Phylogenetic analysis,however, shows them to be members of three distinct lineages.LINK Some mechanisms of self-incompatibility arediscussed in Concept 27.1, pp. 420–421.zoonotic diseases Many infectious pathogens, particularlyviruses, affect individuals of only one species. However, zoonoticdiseases, or zoonoses, are caused by infectious organisms transmittedfrom an infected animal of a different species. Approximately150 zoonotic diseases are known to affect humans; rabies and

320 Chapter 16 | Reconstructing and Using PhylogeniesW.H. Freeman & Company and Sinauer Associates, Inc ©2010Understanding and developing new treatments for many viraldiseases depends in large part on information from phylogeneticanalyses. Different strains of dengue virus require differentforms of treatment, but the virus changes quickly and new variantsare found on a regular basis. Phylogenetic analyses show therelationships of a new dengue variant to known strains, which isan effective way to predict the most effective treatment.Phylogenies allow us to understand the evolutionof complex traitsBiologists are constantly confronted with evolutionary adaptationsthat at first seem puzzling, and many of these have to dowith mating behavior and sexual selection. The tails of male widowbirds(see Figure 15.8) are one of many examples; another isfound among the swordtail fishes.Male swordtail fishes have a long, colorful tail extension (Figure16.8A), and their reproductive success is closely associatedwith this appendage. Males with a long sword are more likelyto mate successfully than are males with a short sword. Severalexplanations have been advanced for the evolution of this structure,including the hypothesis that the sword exploits a preexistingbias in the sensory system of the female fish. This sensoryexploitation hypothesis suggests that female swordtails had a biasto prefer males with long tails even before this trait evolved (perhapsbecause females assess the size of males by their total bodylength—including the tail—and prefer larger males).To test the sensory exploitation hypothesis, phylogeneticanalysis identified the swordtail relatives that had split mostrecently from their lineage before the evolution of sword extensions.These closest relatives turned out to be the platyfishes (Figure16.8B). Even though male platyfishes do not normally haveswords, when researchers attached artificial swordlike structuresto the tails of some male platyfishes, female platyfishes preferredthe males with an artificial sword, thus providing support for thehypothesis that female swordtails had a preexisting sensory biasfavoring tail extensions even before the trait evolved (Figure16.8C). Thus, a long tail became a sexually selected trait becauseof the preexisting preference of the females.FIGURE 16.8 The Origin of aSexually Selected Trait (A) Thelarge tail of male swordtail fishes(genus Xiphophorus) apparentlyevolved through sexual selection,with females mating preferentiallywith males with a longer “sword.”(B) A male platyfish, member of arelated species. (C) Phylogeneticanalysis reveals that the platyfishessplit from the swordtails beforethe evolution of the sword. Theindependent finding that femaleplatyfishes prefer males with anartificial sword further supportsthe idea that this appendageevolved as a result of a preexistingpreference in the females.(A)(B)Ancestral states can be reconstructedIn addition to using phylogenetic methods to infer evolutionaryrelationships, biologists can use these techniques to reconstructthe morphology, behavior, or nucleotide and amino acidsequences of ancestral species (as was demonstrated for the ancestralsequence of bacteriophage T7 in Figure 16.5). For instance,a phylogenetic analysis was used to reconstruct an opsin proteinin the ancestral archosaur (the most recent common ancestor ofbirds, dinosaurs, and crocodiles).Opsins are pigment proteins involved in vision; differentopsins (with different amino acid sequences) are excited bydifferent wavelengths of light (see Figure 35.18). Investigatorsused phylogenetic analysis of opsin from living vertebrates topredict the amino acid sequence of the visual pigment that existedin the ancestral archosaur. A protein with the predictedsequence was then constructed in the laboratory. Investigatorstested the reconstructed opsin and found a significant shift towardthe red end of the spectrum in the light sensitivity of thisprotein compared with that of most modern opsins. Modernspecies that exhibit similar red (and infrared) sensitivity areadapted for nocturnal vision; thus the investigators inferredthat the ancestral archosaur was likely to have been active atnight. These findings may remind you of the movie JurassicPark, although here the extinct species are being “brought backto life” one protein at a time.Molecular clocks help date evolutionary eventsFor many applications, biologists want to know not only theorder in which evolutionary lineages split but also the timingof those splits. In 1965, Emile Zuckerkandl and Linus Paulinghypothesized that rates of molecular change were constantenough that they could be used to predict evolutionary divergencetimes—an idea that has become known as the molecularclock hypothesis.Different genes evolve at different rates. In addition, there aredifferences in evolutionary rates among species related to differinggeneration times, environments, efficiencies of DNA repairsystems, and other biological factors. Nonetheless, among closelyrelated species, a given gene usually evolves at a reasonably(C)Evolution of femalesensory biasEvolution ofmale swordSwordtailfishesPlatyfish

16.3 5.2 Phylogeny Makes Biology Comparative and Predictive 3210.9W.H. Freeman & Company and Sinauer Associates, Inc ©2010Proportion of amino acid differences0.80.70.60.50.40.30.20.10The slope represents anaverage rate of change inamino acid sequences(the molecular clock).0 100 200 300 400 500Time (myr)FIGURE 16.9 A Molecular Clock of the Protein HemoglobinAmino acid replacements in hemoglobin have occurred at a relativelyconstant rate over nearly 500 million years of evolution. Thegraph shows the relationship between time of divergence and proportionof amino acid change for 13 pairs of vertebrate hemoglobinproteins. The average rate of change represents the molecularclock for hemoglobin in vertebrates.constant rate. Therefore, the protein encoded by the gene accumulatesamino acid replacements at a relatively constant rate(Figure 16.9). A molecular clock uses the average rate at whicha given gene or protein accumulates changes to gauge the timeof divergence for a particular split in the phylogeny. Molecularclocks must be calibrated using independent data such as thefossil record (see Concept 18.1), known times of divergence, orbiogeographic dates (e.g., the time of separations of continents;see Concept 18.3). Using such calibrations, times of divergencehave been estimated for many groups of species that have divergedover millions of years.Principles of LIFE Sadava(A) Sinauer AssociatesMorales StudioFigure 16.09 Date 06-16-10Commonancestorof HIV-1(main group)019591990199719831984199419961984199319951983199119831988198719861989198319980.03 0.06 0.09 0.12 0.15 0.18Branch length from common ancestor(B)Branch length from common ancestor0.180.170.160.150.140.130.120.11Besides being useful for dating ancient events, molecularclocks can also be used to study the timing of comparativelyrecent events. Although HIV-1 samples have generally been collectedfrom humans only since the early 1980s, a few isolatesfrom medical biopsies are available from as early as the 1950s.But biologists can use the observed changes in HIV-1 over thepast three decades to project back to the common ancestor of allHIV-1 isolates, and thus can estimate when HIV-1 first enteredhuman populations from chimpanzees (Figure 16.10). Theclock can be calibrated using the samples from the 1980s and1990s, then tested using the samples from the 1950s. As shown inFigure 16.10C, a sample from a 1959 biopsy is dated by molecularclock analysis at 1957 ± 10 years. The molecular clock was alsoused to project back to the common ancestor of this group ofHIV-1 samples. Extrapolation suggests a date of origin for thisgroup of viruses of about 1930. Although AIDS was unknownto Western medicine until the 1980s, this analysis shows thatHIV-1 was present (probably at very low frequency) in humanpopulations in Africa for at least 50 years before its emergenceas a global pandemic. Biologists have used similar analyses toconclude that immunodeficiency viruses have been transmittedrepeatedly into human populations from multiple primates formore than a century.FIGURE 16.10 Dating the Origin of HIV-1 in HumanPopulations (A) A phylo genetic analysis of the main group ofHIV-1 viruses. The dates indicate the years in which samples weretaken. (For clarity, only a small fraction of the samples that wereexamined in the original study are shown.) (B) A plot of year of isolationversus genetic divergence from the common ancestor providesan average rate of divergence, or a molecular clock. (C) Themolecular clock is used to date a sample taken in 1959 (as a test ofthe clock) and the unknown date of origin of the HIV-1 main group(about 1930).Average rate of divergence(molecular clock)0.101980 1985 1990 1995 2000Year(C)0.180.150.120.090.060.03Confidencelimits1959 sample01900 1920 1940 1960 1980 2000YearEstimated date for originof HIV-1 main groupPredictedsampling date1957±10 years

16.4 5.2 Phylogeny Is the Basis of Biological Classification 323evolutionary relationships as the basis for distinguishing, naming,and classifying biological groups.Evolutionary history is the basis for modernbiological classificationW.H. Freeman & Company and Sinauer Associates, Inc ©2010Today’s biological classifications express the evolutionary relationshipsof organisms. Taxa are expected to be monophyletic,meaning that the taxon contains an ancestor and all descendantsof that ancestor, and no other organisms. In other words, a monophyletictaxon is a historical group of related species, or a completebranch on the tree of life. As noted earlier, this is also thedefinition of a clade. A true monophyletic group can be removedfrom a phylogenetic tree by a single “cut” in the tree, as shownin Figure 16.11.Note that there are many monophyletic groups on any phylogenetictree, and that these groups are successively smallersubsets of larger monophyletic groups. This hierarchy of biologicaltaxa, with all of life as the most inclusive taxon and manysmaller taxa within larger taxa, down to the individual species,is the modern basis for biological classification.Although biologists seek to describe and name only monophyletictaxa, the detailed phylogenetic information needed todo so is not always available. A group that does not includeits common ancestor is polyphyletic; a group that does not includeall the descendants of a common ancestor is referred to asparaphyletic (see Figure 16.11). Virtually all taxonomists nowagree that polyphyletic and paraphyletic groups are inappropriateas taxonomic units because they do not correctly reflectevolutionary history. Some classifications still contain suchgroups because some organisms have not been evaluated phylogenetically.As mistakes in prior classifications are detected,taxonomic names are revised and polyphyletic and paraphyleticgroups are eliminated from the classifications.Several codes of biological nomenclature governthe use of scientific namesSeveral sets of explicit rules govern the use of scientific names.Biologists around the world follow these rules voluntarily to facilitatecommunication and dialogue. There may be dozens ofcommon names for an organism in many different languages,and the same common name may refer to more than one species(Figure 16.12). The rules of biological nomenclature aredesigned so that there is only one correct scientific name for anysingle recognized taxon and (ideally) a given scientific nameapplies only to a single taxon (that is, each scientific name isunique). Sometimes the same species is named more than once(when more than one taxonomist has taken up the task); the rulesspecify that the valid name is the first name that was proposed(as in the case of the Leptosiphon/Linanthus genus in Figure 16.6).If the same name is inadvertently given to two different species,then a replacement name must be given to the species that wasnamed second.Because of the historical separation of the fields of zoology,botany (which originally included mycology, the study of fungi),(A) Asclepias tuberosa(B) Castilleja coccinea(C) Hieracium aurantiacumFIGURE 16.12 Same Common Name, Not the SameSpecies All three of these distinct plant species are commonlycalled “Indian paintbrush” in North America. Unique scientificbinomials allow biologists to communicate clearly about each species.(A) Asclepias tuberosa is a perennial milkweed native to easternNorth America. (B) Castilleja coccinea is also native to easternNorth America, but is a member of a very different group of plantscalled scrophs. (C) Hieracium aurantiacum is a European species ofaster that has been widely introduced into North America.Principles of LIFE SadavaSinauer AssociatesMorales StudioFigure 16.12 Date 06-16-10and microbiology, different sets of taxonomic rules were developedfor each of these groups. Yet another set of rules emergedlater for classifying viruses. This separation of fields resulted induplicated taxon names in groups governed by the different setsof rules; Drosophila, for instance, is both a genus of fruit flies anda genus of fungi, and some species in both groups have identicalnames. Until recently these duplicated names caused littleFRONTIERS Biologists are working on a universalcode of nomenclature that can be applied to all organisms,so that every species will have a unique identifying nameor registration number. This will assist efforts to build anonline Encyclopedia of Life that links all the information forall the world’s species.

Lizards324 Chapter 16 | Reconstructing and Using PhylogeniesCrocodilesBirdsApply the ConceptPhylogeny is the basis of biological classificationFrogsSalamandersClassification One:Named groupAmphibiaMammaliaReptiliaAvesIncluded taxaFrogs, salamanders, and caeciliansMammalsTurtles, lizards, and crocodilesBirdsW.H. Freeman & Company and Sinauer Associates, Inc ©2010Consider the above phylogeny and three possible classificationsof the taxa.1. Which of these classifications contains a paraphyleticgroup?Classification One:2. Which of these classifications contains a polyphyleticNamed group? group Included taxaAmphibia Frogs, salamanders, and caeciliansMammalia MammalsReptiliaTurtles, lizards, and crocodilesAvesBirdsconfusion, since traditionally biologists who studied fruit flieswere Classification unlikely to read Two: the literature on fungi (and vice versa).Today, Named given group the prevalence Included of taxa large, universal biological databasesAmphibia (such as GenBank, Frogs, which salamanders, includes and DNA caecilians sequences fromacross Mammalia all life), it is increasingly Mammals important that each taxon haveReptiliaTurtles, lizards, crocodiles, and birdsa unique and unambiguous name.Classification Three:Do You Understand Concept 16.4?Named group Included taxaAmphibia Frogs, salamanders, and caecilians• What Homothermia is difference Mammals between and birds monophyletic, paraphyletic,Reptilia and polyphyletic Turtles, lizards, groups? and crocodiles• Why do biologists prefer monophyletic groups informal classifications?CaeciliansMammals• What advantages or disadvantages do you see tohaving separate sets of taxonomic rules for animals,plants, bacteria, and viruses?TurtlesLizardsCrocodilesBirdsHaving described some of the mechanisms by which evolutionoccurs and how phylogenies can be used to study evolutionaryrelationships, we are now ready to consider the subject of speciation.Speciation is the process that leads to the splitting events(nodes) on the tree of life and eventually results in the millions ofdistinctive species that constitute Earth’s biodiversity.Classification Two:Named groupAmphibiaMammaliaReptiliaClassification Three:Named groupAmphibiaHomothermiaReptiliaIncluded taxaFrogs, salamanders, and caeciliansMammalsTurtles, lizards, crocodiles, and birdsIncluded taxaFrogs, salamanders, and caeciliansMammals and birdsTurtles, lizards, and crocodiles3. Which of these classifications is consistent with the goalof including only monophyletic groups in a biologicalclassification?4. Starting with the classification you named in Question 3,how many additional group names would you need toinclude all the clades shown in this phylogenetic tree?How are phylogenetic methods used toQA resurrect protein sequences from extinctQUESTION organisms in the laboratory?& ANSWERMost genes and proteins of organisms that lived millions of yearsago POL have Hillis decomposed in the fossil remains of these species.Nonetheless, Sinauer Associates sequences of many ancient genes This awaits and proteins a design for can the Apply theMorales StudioConcept featurebe reconstructed Figure APPLY16.02 using Date the 06-15-10 methods described in this chapter.As we discussed in Concept 16.3, just as we can reconstruct themorphological features of a clade’s ancestors, we can also reconstructtheir DNA and protein sequences—if we have enoughinformation about the genomes of their descendants.Biologists have now reconstructed gene sequences from manyspecies that have been extinct for millions of years. Using thisinformation, a laboratory can reconstruct real proteins that correspondto the proteins that were present in long-extinct species.This is how Mikhail Matz and his colleagues were able to resurrectfluorescent proteins from the extinct ancestors of moderncorals and then visualize the colors produced by these proteinsin the laboratory and recreate the probable evolutionary path ofthe different pigments (Figure 16.13).Biologists have even used phylogenetic analysis to reconstructsome protein sequences that were present in the commonancestor of life described in Concept 16.1. These hypothetical