ThegenomeofasongbirdWesleyC.Warren,DavidF.Clayton,HansEllegren,ArthurP - PDF document

ThegenomeofasongbirdWesleyC.Warren,DavidF.Clayton,HansEllegren,ArthurP.Arnold,LaDeanaW.Hillier,AxelKuSteveSearle,SimonWhite,AlbertJ.Vilella,SusanFairley,AndreasHeger,LeshengKong,ChrisP.PontingErichD.Jarvis,ClaudioV.Mello,PatMinx,PeterLovell,TarcisoA.F.Velho,MargaretFerrisChristopherN.Balakrishnan,SaurabhSinha,CharlesBlatti,SarahE.London,YunLi,Ya-ChiLin,JuliaGeorgeJonathanSweedler,BruceSouthey TheGenomeCenter,WashingtonUniversitySchoolofMedicine,CampusBox8501,4444ForestParkAvenue,StLouis,Missouri63108,USA.UniversityofIllinois,Urbana-Champaign,Illinois61801USA. ©2010 generallycharacteristicofbirds),andfeaturesthataremostconspi-cuouslydifferentbetweenthetwolineages—someofwhichwillberelatedtothedistinctiveneuralandbehaviouraltraitsofsongbirds.Wesequencedandassembledamalezebrafinchgenomeusingmethodsdescribedpreviously.Amale(thehomogameticsexinbirds)waschosentomaximizecoverageoftheZchromosome.Ofthe1.2gigabase(Gb)draftassembly,1.0Gbhasbeenassignedto33chromosomesandthreelinkagegroups,byusingzebrafinchgeneticlinkageandbacterialartificialchromosome(BAC)fingerprintmaps.Thegenomeassemblyisofsufficientqualityfortheanalysispresentedhere(seeSupplementaryNote1andSupplementaryTable1).Atotalof17,475protein-codinggeneswerepredictedfromthezebrafinchgenomeassemblyusingtheEnsemblpipelinesupplementedbyGpipegenemodels(SupplementaryNote1).Toextendfurtherthecharacterizationofgenesrelevanttobrainandbehaviour,wealsosequencedcomplementaryDNAsfromtheforebrainofzebrafinchesat50(juvenile,duringthecriticalsonglearningperiod)and850(adult)dayspost-hatch,mappingthesereads(IlluminaGA2)totheprotein-codingmodels(SupplementaryNote1).Ofthe17,475protein-codinggenemodelswefind9,872(56%)and10,106(57%)genesexpressedintheforebrainatthesetwoages(90.7%overlap),respectively.Inadditiontoevidencefordevelopmentalregulation,thesereadsshowfurtherspliceforms,newexonsanduntranslatedsequences(SupplementaryFigs1and2).Toaddressissuesoflarge-scalegenomestructureandevolution,wecomparedthechromosomesofzebrafinchandchickenusingbothsequencealignmentandfluorescentinsituhybridization.Theseana-lysesshowedoverallconservationofsyntenyandkaryotypeinthetwospecies,althoughtherateofintrachromosomalrearrangementwashigh(SupplementaryNote2).Wewerealsosurprisedtoseegenesofthemajorhistocompatibilitycomplex(MHC)dispersedacrossseveralchromosomesinthezebrafinch,incontrasttothesyntenicorganiza-tionofbothchickenandhumanMHCs(SupplementaryNote2).Weassessedspecificgenelossesandexpansionsinthezebrafinchlineagebyconstructingphylogeniesofgenespresentinthelastcommonancestorofbirdsandmammals(SupplementaryNote2andSupplementaryFig.3).Boththezebrafinchandthechickengenomeassemblieslackgenesencodingvomeronasalreceptors,caseinmilkproteins,salivary-associatedproteinsandenamelproteins—notsurprisingly,asbirdslackvomeronasalorgans,mammaryglandsandteeth.Unexpectedly,bothspecieslackthegenefortheneuronalproteinsynapsin1(SYN1);comparativeanalysessuggestthatthelossSYN1andflankinggenesprobablyoccurredinanancestortomodernbirds,possiblywithinthedinosaurlineage(SupplementaryNote2,SupplementaryTable2andSupplementaryFig.4).Bothzebrafinchandchickenhaveextensiverepertoiresofolfactoryreceptor-likesequences(SupplementaryNote2andSupplementaryFig.5),pro-teases(SupplementaryTable3),andarichrepertoireofneuropeptideandpro-hormonegenes.Comparedtomammals,zebrafinchhasduplicationsofgenesencodingseveralproteinswithknownneuralfunctions,includinggrowthhormone,(SupplementaryFig.3),caspase-3and(SupplementaryTable3).Twolargeexpansionsofgenefamiliesexpressedinthebrainseemtohaveoccurredinthezebrafinchlineageafterthesplitfrommammals.Oneinvolvesafamilyrelatedtothe(p21-activatedkinase)gene.Thirty-oneuninterrupted-likesequenceshavebeenidentifiedinthezebrafinchgenome,ofwhich29areexpressedintestisand/orbrain(SupplementaryNote2).Thesecondinvolvesthegene,whichencodesazinc-finger-containingtranscriptionalcontrolprotein.Humansonlyhaveagene,butremarkablythegenehasbeenduplicatedinde-pendently,manytimesinboththezebrafinchandchickenlineagestoformspecies-specificcladesof17and18genes,respectively(Sup-plementaryFig.6).Inthezebrafinchthesegenesareexpressedinthebrain(SupplementaryNote2).Anintriguingpuzzleinaviangenomicshasbeentheevidentlackofachromosome-widedosagecompensationmechanismtobalancetheexpressionofgenesontheZsexchromosome,whichispresentintwocopiesinmalesbutonlyoneinfemales22,23.Thechickenhasbeensuspectedofexertingdosagecompensationonamorelocallevel,bythenon-codingRNAMHM(malehypermethylated)24,25,tocauseacharacteristicvariationofgeneexpressionalongtheZchromosome.Thezebrafinchgenomeassembly,however,lacksanMHMsequence,andgenesadjacenttothecomparableMHMchromosomalpositionshownospecialclusterofdosagecompensation(Fig.1andSup-plementaryNote2).Thus,theputativeMHM-mediatedmechanismofrestrictedZ-chromosomedosagecompensationisnotcommontoallbirds.Chromosomalsexdifferencesinthebraincouldhaveadirectroleinthesexdifferencessoevidentinzebrafinchneuroanatomyandsingingbehaviour.Inmammals,asmuchashalfoftheirgenomesrepresentinter-spersedrepeatsderivedfrommobileelements,whereastheinter-spersedrepeatcontentofthechickengenomeisonly8.5%.Wefindthatthezebrafinchgenomealsohasalowoverallinterspersedrepeatcontent(7.7%),containingalittleover200,000mobileele-ments(SupplementaryTables4and5).Thezebrafinch,however,hasaboutthreetimesasmanyretrovirus-derivedlongterminalrepeat(LTR)elementcopiesasthechicken,andalowcopynumberofshortinterspersedelements(SINEs),whichthechickenlacksaltogether.Expressedsequencetag(EST)analysisshowsthatmobileelementsarepresentinabout4%ofthetranscriptsexpressedinthezebrafinchbrain,andsomeofthesetranscriptsareregulatedbysongexposure(nextsection,Table1).Figure2showsanexampleofanRNAthatwasidentifiedinamicroarrayscreeningforgenesspecificallyenrichedinsongcontrolnucleiandnowseemstorepresentalongnon-codingRNA(ncRNA)containingaCR1-likemobileelement.Theseresultsindicatethatfurtherexperimentsinvestigatingapossibleroleofmobile-element-derivedrepeatedsequencesinvocalcommu-nicationarewarranted.Alargeportionofthegenomeisdirectlyengagedbyvocalcom-munication.ArecentstudydefineddistinctsetsofRNAsinthe MHM * 020,000,000Z chromosome position (bp)40,000,00060,000,000 Figure1Divergentpatternsofdosagecompensationinbirds.a,Themaletofemale(M/F)ratioofgeneexpression,measuredbyspecies-specificmicroarrays,isplottedalongtheZchromosomeofchicken()andzebrafinch().EachpointrepresentstheaverageM/Fratioofaslidingwindowof30genesplottedatthemediangenepositionandsteppingonegeneatatimealongthechromosome.NoteregionoflowerM/FratiosinchickensurroundingthelocusoftheMHM(malehypermethylated)ncRNA.Inzebrafinch,genesadjacenttothecomparableMHMposition(asterisk)shownospecialclusterofdosagecompensation(lowM/Fratios),andnoMHMsequenceappearsinthegenomeassembly.bp,basepairs.LETTERSVol4641April2010©2010 auditoryforebrainthatrespondindifferentwaystosongplaybacksduringtheprocessofsong-specifichabituation,aformoflearningWenowmapeachofthesesong-responsiveRNAstothegenomeassembly(Table1andSupplementaryNote3).Notably,wefindevidencethat40%oftranscriptsintheunstimulatedauditoryfore-brainarenon-codingandderivefromintronicorintergenicloci(Table1).AmongtheRNAsthatarerapidlysuppressedinresponsetonewvocalsignals(‘noveldown’),two-thirdsarencRNAs.TherobustinvolvementofncRNAsintheresponsetosongledustoaskwhethersongexposurealterstheexpressionofmicroRNAs—smallncRNAsthatregulategeneexpressionbybindingtotargetmessengerRNAs.IndeedwefindthatmiR-124,aconservedmicroRNAimplicatedinneurologicalfunctioninotherspecies,israpidlysuppressedinresponsetosongplaybacks(Fig.3).Weinde-pendentlymeasuredthiseffectbydirectIlluminasequencingofsmallRNAsintheauditoryforebrain,andalsoidentifiedotherknownandnewmicroRNAs,severalofwhichalsochangeinexpressionaftersongstimulation(SupplementaryNote2).ApotentialsiteofactionformicroRNAswasshownbygenomicmappingoftranscriptsthatincreaserapidlyafternewsongexposure(Table1,‘novelup’).TwoofthecDNAclonesthatmeasuredthemostrobustincreasesaligntoanunusuallylong(3kilobases(kb))untranslatedregion(UTR)inthehumangenethatencodestheNR4A3transcriptionfactorprotein(Fig.4a).TheentireUTRissimilarinhumansandzebrafinches,withseverallongsegmentsof80%identity(Fig.4b).Withinthesesegmentswefindconservedpredictedbindingsitesfor11differentmicroRNAs,includingfivenewmicroRNAsfoundbydirectsequencingofsmallRNAsfromthezebrafinchforebrain(Fig.4b).Thesefindingsindicatethatthistranscriptelementmayfunctioninbothhumansandsong-birdstointegratemanyconservedmicroRNAregulatorypathways.Theactofsingingalsoaltersgeneexpressioninsongcontrol,andweusedthegenomeassemblytoanalysethetranscrip-tionalcontrolstructureofthisresponse.Usingoligonucleotidemicroarrays,weidentified807genesinwhichexpressionsignifi-cantlychangedasaresultofsinging.Theseweregroupedbyclusteringinto20distinctexpressionprofileclusters(Fig.5aandSupplementaryNote3).Generegulatorysequences(transcription-factor-bindingsites)werepredictedacrossthegenomeusinganewmotif-scanningapproach(SupplementaryNote1),andweobservedsignificantcorrelationbetweenchangesinexpressionoftranscrip-tionfactorgenesandtheirpredictedtargets(Fig.5bandSup-plementaryTable6).Thus,theexperienceofsingingandhearingsongengagescomplexgeneregulatorynetworksintheforebrain,alteringtheexpressionofmicroRNAs,transcriptionfactorgenes,andtheirtargets,aswellasofnon-codingRNAelementsthatmayintegratetranscriptionalandpost-transcriptionalcontrolsystems.Learnedvocalcommunicationiscrucialtothereproductivesuccessofasongbird,andthisbehaviourevolvedafterdivergence Table1StructuralfeaturesofthesongresponsivegenomeAllgenesanalysedNovelupNoveldownHabituateupHabituatedownAllESTsMappedlociEnsemblgenesMobileelementcontentNumberwithmobileelementsPercentagemobileelements41922Codingandnon-codingcontentmRNAtranscripts(%(-value))ESTlocimappedtointrons(%(Intergenicloci(%(Protein-codinggeneterritoriesMeangenelength(kb)Intergeniclength(kb)Territorysize(kb)Amicroarraymadefromnon-redundantbrain-derivedESTswasusedtodefinefoursubgroupsofRNAsthatshowdifferentresponsesinauditoryforebraintosongexposures(novelupanddown,habituatedupanddown).TheseESTsweremappedtogenomepositionsasdescribed(SupplementaryNote3).AllESTswereanalysedformobileelementcontentusingRepeatMasker(SupplementaryNote2).-valueisforthecomparisontoallgenes(Fisher’sexacttest).AllESTsthatcouldbemappeduniquelytothegenomeassemblywereassessedforoverlapwithEnsemblannotationsofmRNAtranscripts(proteincodingandUTRs),intronicregions,orintergenic-valueisforcomparisontoallmappedloci(Fisher’sexacttest).Resultsarethepercentagewithvaluesinparentheseswhereshown.Thesizeofeachuniqueprotein-codinggeneterritorywasdeterminedbycombiningthelengthoftheEnsemblgenemodelwithitsintergenicspacing.Th-valueisforthecomparisontoallgenes,usingatwo-tailedWilcoxonranksumtest. cdMeso.Stb dHpShelf HVCLMANCbStNidoMeso c DV949717CR1-like elementFE733533EE061829Chr 374,593,50074,594,50074,595,500a Figure2EnrichedexpressionofaCR1-likeelementinthezebrafinchsongsystem.a,GenomicalignmentofanRNAcontainingaCR1-likeretrotransposonelement(inblue)andadjacentESTs,withrespectiveGenBankaccessionnumbers.,DV949717isexpressedinthebrainofadultmaleswithenrichmentinsongnucleiHVC(letter-basedname)andLMAN(lateralmagnocellularnucleusoftheanteriornidopallium),asrevealedbyinsituhybridization.Thediagraminindicatesareasshowninphotomicrographsin.Cb,cerebellum;Hp,hippocampus;Meso,mesopallium;Nido,nidopallium;Shelf,nidopallialshelfregion;St,striatum.Scalebars,0.1mm. 0.00 0.25 0.50 0.75 1.00 1.25 Relative gene expression Figure3miR-124intheauditoryforebrainissuppressedbyexposuretonewsong.TaqManassayscomparingsamplesfromtheauditorylobuleofadultmalezebrafinchesinsilence(openbars)or30minafteronsetofnewsongplayback(filledbars).,Comparisonoftwosamplepools,eachcontainingauditoryforebrainsof20birds.,Comparisonsofpairedindividualsubjects,6pairs(0.03,Wilcoxonpairedtest).Errorbarsdenotes.e.m.oftriplicateTaqManassays.ParallelTaqMananalysesofthesmallRNAwereperformedwithallsamplesandshowednosignificanteffectoftreatmentforthiscontrolRNA.Vol4641April2010LETTERS©2010 ofthesongbirdlineage.Thus,itseemslikelythatgenesinvolvedintheneurobiologyofvocalcommunicationhavebeeninfluencedbypositiveselectioninsongbirds.Withthisinmind,weexaminedtheintersectionoftwosetsofgenes:(1)thosethatrespondtosongexposureintheauditoryforebrainasdiscussedintheprevioussec-tion;and(2)thosethatcontainresiduesthatseemtohavebeenpositivelyselectedinthezebrafinchlineage,asdeterminedusingphylogeneticanalysisbymaximumlikelihood(PAML)(Supplemen-taryNote4).Thereare214genesthatarecommontobothlists.Ofthese,49aresuppressedbysongexposure(SupplementaryTable7),and6ofthese49areexplicitlyannotatedforionchannelactivity(Table2).Thisyieldsahighlysignificantstatisticalenrichmentfortheterm‘ionchannelactivity’(0.0016,falsediscoveryrate(FDR)adjustedFisher’sexacttest)andotherrelatedtermsinthissubsetofgenes(SupplementaryTables8and9).Independentevidencehasalsodemonstrateddifferentialanatomicalexpressionofionchannelgenesinsongcontrolnuclei.Ionchannelgeneshaveimportantrolesinmanyaspectsofbehaviour,neurologicalfunctionand.Thisclassofgenesishighlylikelytobelinkedtosongbehaviourandshouldbeamajortargetforfuturefunctionalstudies.Passerinesrepresentoneofthemostsuccessfulandcomplexradia-tionsofterrestrialanimals.Herewepresentthefirst,toourknow-ledge,analysisofthegenomeofapasserinebird.Thezebrafinchwaschosenbecauseofitswell-developedstatusasamodelorganismforanumberoffieldsinbiology,includingneurobiology,ethology,ecology,biogeographyandevolution.Inthezebrafinchasinthechicken,weseeasmaller,tightergenomecomparedtomammals,withamarkedreductionofinterspersedrepeats.Thezebrafinchpresentsapictureofgreatergenomicplasticitythanmighthavebeenexpectedfromthechickenandotherprecedents,withahighdegreeofintrachromosomalrearrangementsbetweenthetwoavianspecies,genecopynumbervariationsandtranscribedmobileelements.Yetwealsoseeanoverallsimilaritytomammalsinprotein-codinggenecontentandcoretran-scriptionalcontrolsystems.Ouranalysissuggestsseveralchannelsthroughwhichevolutionmayhaveactedtoproducetheuniqueneurobiologicalpropertiesofsongbirdscomparedtothechickenandotheranimals.Theseincludethemanagementofsexchromosomegeneexpression,acceleratedevolutionofneuronaliontransportgenes,geneduplicationstopro-ducenewvariantsofandotherneurobiologically IntergeneIntron80%77%88%81%mir-17-5pmir-17-5p Figure4UTRisapotentialregionformicroRNAintegration.a,zPicturealignmentof3portionofzebrafinchtohumanshowingUTRregionofhighsimilaritybeyondthecodingexons.Darkredbars,regionswiththehighestsequenceconservation;blackrectangles,positionofsong-regulatedESTswithintheconservedUTRbutoutsidetheEnsemblgenemodel(ENSTGUG00000008853).,Alignmentofzebrafinchandhuman3UTRsequencesshowingthepercentsequenceidentityforeachevolutionarilyconservedregion.Dotsindicatepositionsofconservednew(‘n-’)orestablished(‘miR-’)microRNA-bindingsitesinbothspecieswithintheseregions. ab 00.51234ATF1ATF4ÐATFATFOver-represented binding motifsLog(fold expression change) Figure5TranscriptionalcontrolnetworkinareaXengagedbysinging.,Clustered(1–20)temporalexpressionprofilesof807genes(rows)thatchangewithtimeandamountofsinging;red,increases;blue,decreases;white,nochangerelativetoaverage0-hcontrol.Grey/colouredbarsonleft,clusterswithenrichmentofspecificpromotermotifs(,Enrichedtranscription-factor-bindingmotifs(abbreviations)foundinthepromotersoflateresponsegenes,clusters9–12(colouredasin);bold,bindingsitesforknownactivity-dependenttranscriptionfactors(forexample,CREBP1)ortranscriptionfactorcomplexes(forexample,CREBP1–CJUN);black,sitesforpost-translationallyactivatedtranscriptionfactors;brown,sitesfortranscriptionallyactivatedtranscriptionfactorsincludingbysinging(forexample,incluster1).Graphshowstimecourseofaverageexpressionofallgenesinthelateresponseclusters,normalizedtoaverage0hforthatcluster.AlsoplottedistheaverageexpressionofthetranscriptionfactormRNA,whichbindstotheAP-1siteover-representedinthepromotersofcluster10genes.LETTERSVol4641April2010©2010 importantgenes,andanewarrangementofMHCgenes.Mostnotably,ouranalysessuggestalargerecruitmentofthegenomeduringvocalcommunication,includingtheextensiveinvolvementofncRNAs.IthasbeenproposedthatncRNAshaveacontributingroleinenablingordrivingtheevolutionofgreatercomplexityinhumansandothercomplexeukaryotes.Seeingthatlearnedvocalcommunicationitselfisaphenomenonthathasemergedonlyinsomeofthemostcomplexorganisms,perhapsncRNAsareanexusofthisphenomenon.Muchworkwillbeneededtoestablishtheactualfunctionalsig-nificanceofmanyoftheseobservationsandtodeterminewhentheyaroseinavianevolution.ThisworkcannowbeexpeditedwiththerecentdevelopmentofamethodfortransgenesisinthezebrafinchAnimportantgenerallesson,however,isthatdynamicandseren-dipitousaspectsofthegenomemayhaveunexpectedrolesintheelaboratevocalcommunicativecapabilitiesofsongbirds.METHODSSUMMARYSequenceassembly.Sequencedreadswereassembledandattemptsweremadetoassignthelargestcontiguousblocksofsequencetochromosomesusingageneticlinkagemap,fingerprintmapandsyntenywiththechickengenomeassemblyGallus_gallus-2.1,arevisedversionoftheoriginaldraft(SupplementaryNote1).GeneorthologyassignmentwasperformedusingtheEnsemblComparaGeneTreespipelineandtheOPTICpipeline(SupplementaryNote1).OrthologyrateestimationwasperformedwithPAML(pairwisemodel0,Nssites0).Inallcases,codonfrequencieswereestimatedfromthenucleotidecompositionateachcodonposition(F3X4model).Geneexpressionandevolution.MethodsforIlluminareadcounting,insituhybridization,TaqManRT–PCR,microarrays,regulatorymotifandevolution-aryrateanalysesaregiveninSupplementaryNotes1–4.Received30September2009;accepted6January2010.1.Zann,R.A.TheZebraFinch:ASynthesisofFieldandLaboratoryStudiesUniv,Press,1996).2.Clayton,D.F.,Balakrishnan,C.N.&London,S.E.Integratinggenomes,brainandbehaviorinthestudyofsongbirds.Curr.Biol.19,R873(2009).3.Nottebohm,F.inHopeForaNewNeurology(ed.Nottebohm,F.)(NewYorkAcademyofScience,1985).4.Doupe,A.J.&Kuhl,P.K.Birdsongandhumanspeech:commonthemesandAnnu.Rev.Neurosci.631(1999).5.Jarvis,E.D.Learnedbirdsongandtheneurobiologyofhumanlanguage.Ann.NYAcad.Sci.777(2004).6.Hillier,L.W.etal.Sequenceandcomparativeanalysisofthechickengenomeprovideuniqueperspectivesonvertebrateevolution.7.Hackett,S.J.etal.Aphylogenomicstudyofbirdsrevealstheirevolutionary1768(2008).8.Zeigler,H.P.&Marler,P.BehavioralNeurobiologyofBirdSongVol.1016(NewYorkAcademyofSciences,2004).9.Hahnloser,R.H.,Kozhevnikov,A.A.&Fee,M.S.Anultra-sparsecodeunderliesthegenerationofneuralsequencesinasongbird.10.Mooney,R.Neuralmechanismsforlearnedbirdsong.Learn.Mem.11.Konishi,M.&Akutagawa,E.Neuronalgrowth,atrophyanddeathinasexuallydimorphicsongnucleusinthezebrafinchbrain.147(1985).12.Goldman,S.A.&Nottebohm,F.Neuronalproduction,migration,anddifferentiationinavocalcontrolnucleusoftheadultfemalecanarybrain.NatlAcad.Sci.USA2394(1983).13.Nottebohm,F.Theroadwetravelled:discovery,choreography,andsignificanceofbrainreplaceableneurons.Ann.NYAcad.Sci.658(2004).14.London,S.E.,Remage-Healey,L.&Schlinger,B.A.Neurosteroidproductioninthesongbirdbrain:Are-evaluationofcoreprinciples.Front.Neuroendocrinol.314(2009).15.Mello,C.V.,Vicario,D.S.&Clayton,D.F.Songpresentationinducesgeneexpressioninthesongbirdforebrain.Proc.NatlAcad.Sci.USA89,68186822(1992).16.Dong,S.&Clayton,D.F.Habituationinsongbirds.Neurobiol.Learn.Mem.188(2009).17.Woolley,S.C.&Doupe,A.J.Socialcontext-inducedsongvariationaffectsfemalebehaviorandgeneexpression.PLoSBiol.e62(2008).18.Jarvis,E.D.,Scharff,C.,Grossman,M.R.,Ramos,J.A.&Nottebohm,F.Forwhomthebirdsings:context-dependentgeneexpression.788(1998).19.Clayton,D.F.Thegenomicactionpotential.Neurobiol.Learn.Mem.20.Warren,W.C.etal.Genomeanalysisoftheplatypusrevealsuniquesignaturesof183(2008).21.Stapley,J.,Birkhead,T.R.,Burke,T.&Slate,J.AlinkagemapofthezebrafinchTaeniopygiaguttataprovidesnewinsightsintoaviangenomeevolution.667(2008).22.Itoh,Y.etal.Dosagecompensationislesseffectiveinbirdsthaninmammals.2(2007).23.Ellegren,H.etal.Facedwithinequality:chickendonothaveageneraldosagecompensationofsex-linkedgenes.BMCBiol.40(2007).24.Teranishi,M.etal.TranscriptsoftheMHMregiononthechickenZchromosomeaccumulateasnon-codingRNAinthenucleusoffemalecellsadjacenttotheChromosomeRes.165(2001).25.Arnold,A.P.,Itoh,Y.&Melamed,E.Abird’s-eyeviewofsexchromosomedosageAnnu.Rev.GenomicsHum.Genet.127(2008).26.Lovell,P.V.,Clayton,D.F.,Replogle,K.L.&Mello,C.V.Birdsong‘‘transcriptomics’’:neurochemicalspecializationsoftheoscinesongsystem.e3440(2008).27.Dong,S.etal.Discretemolecularstatesinthebrainaccompanychangingresponsestoavocalsignal.Proc.NatlAcad.Sci.USA11369(2009).28.Makeyev,E.V.&Maniatis,T.Multilevelregulationofgeneexpressionby1790(2008).29.Wada,K.etal.Amolecularneuroethologicalapproachforidentifyingandcharacterizingacascadeofbehaviorallyregulatedgenes.Proc.NatlAcad.Sci.USA15217(2006).30.Wada,K.,Sakaguchi,H.,Jarvis,E.D.&Hagiwara,M.Differentialexpressionofglutamatereceptorsinavianneuralpathwaysforlearnedvocalization.J.Comp.64(2004).31.Cooper,E.C.&Jan,L.Y.Ionchannelgenesandhumanneurologicaldisease:recentprogress,prospects,andchallenges.Proc.NatlAcad.Sci.USA4766(1999).32.Mattick,J.S.RNAregulation:anewgenetics?NatureRev.Genet.33.Agate,R.J.,Scott,B.B.,Haripal,B.,Lois,C.&Nottebohm,F.Transgenicsongbirdsofferanopportunitytodevelopageneticmodelforvocallearning.Proc.NatlAcad.Sci.USA17967(2009).34.Replogle,K.etal.TheSongbirdNeurogenomics(SoNG)Initiative:community-basedtoolsandstrategiesforstudyofbraingenefunctionandevolution.131(2008).35.Ovcharenko,I.,Loots,G.G.,Hardison,R.C.,Miller,W.&Stubbs,L.zPicture:dynamicalignmentandvisualizationtoolforanalyzingconservationprofiles.GenomeRes.477(2004).SupplementaryInformationislinkedtotheonlineversionofthepaperatwww.nature.com/nature.AcknowledgementsThesequencingofzebrafinchwasfundedbytheNationalHumanGenomeResearchInstitute(NHGRI).FurtherresearchsupportincludedgrantstoD.F.C.(NIHRO1NS045264andRO1NS051820),H.E.(SwedishResearchCouncilandKnutandAliceWallenbergFoundation),E.D.J.(HHMI,NIHDirectorsPioneerAwardandR01DC007218),M.A.B.(NIHRO1GM59290)andJ.S.(BiotechnologyandBiologicalSciencesResearchCouncilgrantnumberBBE0175091).Resourcesforexploringthesequenceandannotationdataare Table2Song-suppressedionchannelgenesunderpositiveselectionSitesPS/totalVoltage-dependentN-typecalciumchannelsubunitVoltage-dependentT-typecalciumchannelsubunitGlutamatereceptorprecursor(GluR-,AMPAGlutamatereceptorprecursor(GluR-,AMPAPotassiumvoltage-gatedchannelsubfamilyCmemberTransientreceptorpotentialcationchannelsubfamilyVmemberThesesixgenesaresuppressedbysongexposure(FDRandtheyshowevidenceofpositiveselectioninthezebrafinchrelativetochicken(,SupplementaryNote3).Branchdenotesthedifferenceinthenon-synonymoustosynonymoussubstitutionratio(d)betweenzebrafinchandotherbirds(chickenandtheancestralbranchleadingtochickenandzebrafinch).Positivevaluesindicatethatthegeneisrapidlyevolving,whereasnegativevaluesindicategenesevolvingmoreslowly.SitesPS/totaldenotesthenumberofindividualsiteswithempiricalBayesposteriorprobabilitygreaterthan0.95of1(positiveselection)inthefinchversusthetotalnumberofresiduesintheprotein,frombranch-sitemodelanalysisimplementedinPAML.Notethatgenescanshowoverallslowerevolutioninthebranchmodelyetshowevidenceofsignificantpositiveselectionatspecificsites.Gene-widedifferencesthatweresignificant(0.05)byalikelihoodratiotest.Vol4641April2010LETTERS©2010 availableonbrowserdisplaysavailableatUCSC(http://genome.ucsc.edu),Ensembl(http://www.ensembl.org),theNCBI(http://www.ncbi.nlm.nih.gov)andhttp://aviangenomes.org.WethankK.Lindblad-Tohforpermissiontousethegreenanolelizardgenomeassembly,theProductionSequencingGroupofTheGenomeCenteratWashingtonUniversitySchoolofMedicineforgeneratingallthesequencereadsusedforgenomeassembly,andtheClemsonUniversityGenomeInstitutefortheconstructionoftheBAClibrary.Wewouldliketorecognizealltheimportantpublishedworkthatwewereunabletociteowingtospacelimitations.AuthorContributionsW.C.W.,D.F.C.,H.E.andA.P.A.comprisetheorganizingcommitteeofthezebrafinchgenomesequencingproject.Projectplanning,managementanddataanalysis:W.C.W.,D.F.C.,H.E.andA.P.A.Assemblyannotationandanalysis:L.W.H.,P.M.,S.-P.Y.,L.Y.,J.N.,A.C.,S.H.,J.Sl.,J.St.,D.B.andS.-P.Y.Proteincodingandnon-codinggeneprediction:S.S.,C.B.,P.F.,S.W.,A.H.,C.P.P.andL.K.SNPanalysis:P.F.andW.M.M.Orthologypredictionandanalysis:A.J.V.,A.H.,C.P.P.,S.F.andL.K.Repeatelementanalysis:M.A.B.,A.F.A.S.,R.H.,M.K.K.,J.A.W.,W.G.andD.D.P.Segmentalduplicationandgeneduplicationanalysis:L.C.,Z.C.,E.E.E.,L.K.,C.P.P.,M.F.,C.N.B.,R.E.,J.G.andS.E.L.Proteaseannotationandanalysis:X.S.P.,V.Q.,G.V.andC.L.-O.Neuropeptidehormoneannotation:J.Sw.andB.S.Smallnon-codingRNAanalysis:Y-C.L.,Y.L.,P.G.,M.W.andX.L.Comparativemapping:D.K.G.,M.V.andB.M.S.Singinginducedgenenetworkanalysis:E.D.J.,A.R.P.,O.W.andJ.H.Z-chromosomeanalysis:Y.I.andA.P.A.Geneexpressionandinsituanalysisandsynapsinsynteny/lossanalysis:C.V.M.,P.L.andT.A.F.V.Adaptiveevolutionanalysis:A.K.,K.N.,N.B.,L.S.,B.N.andC.N.B.Geneexpressioninthebrainanalysis:C.S.,I.A.,A.S.,H.L.,H.R.andM.S.MHCanalysis:S.E.,C.N.B.andR.E.Olfactoryreceptoranalysis:T.O.,D.L.andL.K.Sequencingmanagement:R.K.W.,E.R.M.andL.F.Physicalmapconstruction:T.G.Zebrafinchtissueresources:T.Bu.andT.Bi.ZebrafinchcDNAresources:D.F.C.,E.D.J.andX.L.AuthorInformationTaeniopygiaguttatawhole-genomeshotgunprojecthasbeendepositedinDDBJ/EMBL/GenBankundertheprojectaccessionABQF00000000.Reprintsandpermissionsinformationisavailableatwww.nature.com/reprints.ThispaperisdistributedunderthetermsoftheCreativeCommonsAttribution-Non-Commercial-ShareAlikelicence,andisfreelyavailabletoallreadersatwww.nature.com/nature.Theauthorsdeclarenocompetingfinancialinterests.CorrespondenceandrequestsformaterialsshouldbeaddressedtoW.C.W.(wwarren@watson.wustl.edu),D.F.C(dclayton@illinois.edu),H.E.(hans.ellegren@ebc.uu.se)orA.P.A.(arnold@ucla.edu). LETTERSVol4641April2010©2010