A chromosome-level genome of Astyanax mexicanus surface ...

The improved contig length of the surface fish genome and our syntenic analysis that unites the physical genome to a previously published ... Skiptomaincontent Thankyouforvisitingnature.com.YouareusingabrowserversionwithlimitedsupportforCSS.Toobtain thebestexperience,werecommendyouuseamoreuptodatebrowser(orturnoffcompatibilitymodein InternetExplorer).Inthemeantime,toensurecontinuedsupport,wearedisplayingthesitewithoutstyles andJavaScript. Advertisement nature naturecommunications articles article Achromosome-levelgenomeofAstyanaxmexicanussurfacefishforcomparingpopulation-specificgeneticdifferencescontributingtotraitevolution DownloadPDF Subjects ComparativegenomicsEvolutionarygenetics AbstractIdentifyingthegeneticfactorsthatunderliecomplextraitsiscentraltounderstandingthemechanisticunderpinningsofevolution.Cave-dwellingAstyanaxmexicanuspopulationsarewelladaptedtosubterraneanlifeandmanypopulationsappeartohaveevolvedtroglomorphictraitsindependently,whilethesurface-dwellingpopulationscanbeusedasaproxyfortheancestralform.Herewepresentahigh-resolution,chromosome-levelsurfacefishgenome,enablingthefirstgenome-widecomparisonbetweensurfacefishandcavefishpopulations.Usingthisresource,weperformedquantitativetraitlocus(QTL)mappinganalysesandfoundnewcandidategenesforeyelosssuchasdusp26.WeusedCRISPRgeneeditinginA.mexicanustoconfirmtheessentialroleofagenewithinaneyesizeQTL,rx3,ineyeformation.Wealsogeneratedthefirstgenome-wideevaluationofdeletionvariabilityacrosscavefishpopulationstogaininsightintothispotentialsourceofcaveadaptation.Thesurfacefishgenomereferencenowprovidesamorecompleteresourceforcomparative,functionalandgeneticstudiesofdrastictraitdifferenceswithinaspecies. DownloadPDF IntroductionEstablishingtherelationshipbetweennaturalenvironmentalfactorsandthegeneticbasisoftraitevolutionhasbeenchallenging.Theecologicalshiftfromsurface-to-caveenvironmentsprovidesatractablesystemtoaddressthis,asinthisinstancethepolarityofevolutionarychangeisknown.Acrosstheglobe,subterraneananimals,includingfish,salamanders,rodents,andmyriadsofinvertebratespecies,haveconvergedonreductionsinmetabolicrate,eyesize,andpigmentation1,2,3.Therobustphenotypicdifferencesfromsurfacerelativesprovidetheopportunitytoinvestigatethemechanisticunderpinningsofevolutionandtodeterminewhethergenotype–phenotypeinteractionsaredeeplyconserved.TheMexicancavefish,Astyanaxmexicanus,hasemergedasapowerfulmodeltoinvestigatecomplextraitevolution4.A.mexicanuscomprisesatleast30cave-dwellingpopulationsintheSierradeElAbra,SierradeColmena,andSierradeGuatemalaregionsofnortheasternMexico,andsurface-dwellingfishofthesamespeciesinhabitriversandlakesthroughoutMexicoandsouthernTexas5.Theecologyofsurfaceandcaveenvironmentsdifferdramatically,allowingforfunctionalandgenomiccomparisonsbetweenpopulationsthathaveevolvedindistinctenvironments.Dozensofevolvedtraitdifferenceshavebeenidentifiedincavefishincludingchangesinmorphology,physiology,andbehavior4,6.Itisalsoworthnotingthatcomparingcavefishtosurfacefishrevealssubstantialdifferencesinmanytraitsofpossiblerelevancetohumandisease,includingsleepduration,circadianrhythmicity,anxiety,aggression,heartregeneration,eyeandretinadevelopment,craniofacialstructure,insulinresistance,appetite,andobesity7,8,9,10,11,12,13,14,15,16,17.Further,generationoffertilesurface-cavehybridsinalaboratorysettinghasallowedforgeneticmappinginthesefish10,18,19,20,21,22,23,24,25.Clearphenotypicdifferences,combinedwithavailabilityofgenetictools,positionsA.mexicanusasanaturalmodelsystemforidentifyingthegeneticbasisofecologicallyandevolutionaryrelevantphenotypes17,26,27.Recently,thegenomeofanindividualfromthePachóncavepopulationwassequenced28.Whilethisworkuncoveredgenomicintervalsandcandidategeneslinkedtocavetraits,animportantcomputationallimitationwassequencefragmentationduetotheuseofshort-readsequencing.Inaddition,lackofasurfacefishgenomepreventsdirectcomparisonsbetweensurfaceandcavefishpopulationsatthegenomiclevel.Here,weaddressthesetwoobstaclesbypresentingthefirstdenovogenomeassemblyofthesurfacefishmorphusinglong-readsequencingtechnology.Thisapproachyieldedamuchmorecomprehensivegenomethatallowsfordirectgenome-widecomparisonsbetweensurfacefishandPachóncavefish.Asaproofofprinciple,wefirstconfirmedknowngeneticmutationsassociatedwithpigmentationandeyeloss,thendiscoverednovelquantitativetraitloci(QTL),andidentifiedcodinganddeletionmutationsthathighlightputativecontributionstocavetraitbiology.ResultsWesetouttogeneratearobustreferencegenomeforthesurfacemorphofA.mexicanusfromasinglelab-rearedfemale,descendedfromwild-caughtindividualsfromknownMexicolocalities(Fig. 1a,SupplementaryFig. 1).WesequencedandassembledthegenomeusingPacificBiosciencessingle-moleculereal-time(SMRT)sequencing(~73×genomecoverage)andthewtdbg2assembler29toanungappedsizeof1.29 Gb.InitialscaffoldingofassembledcontigswasaccomplishedwiththeaidofanA.mexicanussurfacefishphysicalmap(BioNano)followedbymanualassignmentof70%oftheassemblyscaffoldsto25totalchromosomesusingtheexistingA.mexicanusgeneticlinkagemapmarkers30.Thefinalgenomeassembly,Astyanaxmexicanus2.0,comprisesatotalof2415scaffolds(includingsinglecontigscaffolds)withN50contigandscaffoldlengthsof1.7and35 Mb,respectively,whichiscomparabletoothersimilarlysequencedandassembledteleostfishes(SupplementaryTable 1).Theassembledregions(394 Mb)thatwewereunabletoassigntochromosomesweremostlydueto3.36%(2235markerstotal)ofthegeneticlinkagemarkersnotaligningtothesurfacefishgenome,singlemarkerspercontigwheretheorientationcouldnotbeproperlyassigned,ormarkersthatmappedtomultipleplacesinthegenomeandthuscouldnotbeuniquelymapped.Theuniquelymappedmarkersexhibitedfeworderingdiscrepanciesandsignificantsyntenybetweenthelinkagemap30andtheassembledscaffoldsoftheAstyanaxmexicanus2.0genome,validatingtheorderandorientationofamajorityoftheassembly(Fig. 1b;SupplementaryFig. 2).InAstyanaxmexicanus2.0,weassembleandidentify11%moretotalmaskedrepeatsthanintheAstyanaxmexicanus1.0.2assembly(SupplementaryTable 1).Amongstasmallsamplingofassembledteleostgenomes,A.mexicanusappearstobeanintermediateforestimatesoftotalinterspersedrepeatsusingWindowMasker31at41%,comparedtoXiphophorusmaculatus(27%)andDaniorerio(50%).Possiblemappingbiasacrosscavepopulationsequencestothecavevssurfacefishgenomereferenceswasalsoinvestigatedbymappingpopulationlevelresequencingreadstobothgenomes32.WefoundthenumberofunmappedreadsisgreaterforallpopulationsalignedtotheAstyanaxmexicanus1.0.2referencecomparedtoAstyanaxmexicanus2.0(Fig. 1c).Also,thepercentageofproperlypairedreads,thatis,pairswherebothendsaligntothesamescaffold,isgreaterforallresequencedcavepopulationsalignedtotheAstyanaxmexicanus2.0reference(SupplementaryFig. 3)andsignificantlymorenonprimaryalignmentswithgreatervariationwereobserved(SupplementaryFig. 4).BothmetricsindicatethattheAstyanaxmexicanus2.0referencehasmoreresolvedsequenceregionsthantheAstyanaxmexicanus1.0.2reference.FutureapplicationofphasedassemblyapproacheswilllikelyresolveasignificantlyhigherproportionofchromosomalsequencesintheA.mexicanusgenome33.Fig.1:AssemblymetricsforthesurfacefishgenomeAstyanaxmexicanus2.0.aAdultAstyanaxmexicanussurfacefish.bSubstantialsyntenyisevidentbetweenarecombination-basedlinkagemap30andthedraftsurfacefishgenome.Bymappingtherelativepositionsofgenotyping-by-sequencing(GBS)markersfromadensemapconstructedfromaPachón × surfacefishF2pedigree,weobservedsignificantsyntenybasedon96.6%ofourgenotypemarkers.cProportionofunmappedreadsforthesamesamplesalignedtothecave(squarepoints)andsurface(trianglepoints)Astyanaxreferencegenomes.Colorsunitesamplesbypopulationidentity.Bestfitlinesforeachalignment(cave:blue,surface:yellow)havesimilarslope,indicatingthatpopulationidentityhasasimilareffectonmappingratetoeithergenome.FullsizeimageTwoindependentsetsofprotein-codinggenesweregeneratedusingtheNCBI34andEnsembl35automatedpipelineswithsimilarnumbersofgenesfoundbyeach:25,293and26,698,respectively(SupplementaryTable 2).Geneannotationwasaidedbythediversityoftranscriptdataderivedfromwholeadultfish,embryos,and12differenttissuesavailablefromtheNCBIshort-readarchive.Asaresult,thetotalpredictedprotein-codinggenesandtranscripts(mRNA)wereconsistentwithotherannotatedteleostspecies(SupplementaryTable 2)and1665newprotein-codinggeneswereaddedcomparedtoAstyanaxmexicanus1.0.2.Inaddition,longnoncodinggene(e.g.,lncRNA)representationissignificantlyimprovedintheAstyanaxmexicanus2.0referencecomparedtotheAstyanaxmexicanus1.0.2reference(5314vs1062),althoughtargetednoncodingRNAsequencingwillberequiredtoachieveannotationcomparabletozebrafish(SupplementaryTable 2).Weassessedcompletenessofgeneannotationbyapplyingbenchmarkinguniversalsingle-copyorthologscores,whichmeasuregenecompletenessamongvertebrategenes.Wefind94.6%ofgenesarecomplete,4%aremissing,and4.2%areduplicated(SupplementaryTable 3).Inaddition,usingNCBIstranscriptalignerSplign(C++toolkit)geneannotationmetrics,same-speciesRefSeqorGeneBanktranscriptsshow98.4%coveragewhenalignedtoAstyanaxmexicanus2.0.Intotal,ourmeasuresofgenerepresentationintheAstyanaxmexicanus2.0referenceshowahigh-qualityresourceforthestudyofA.mexicanusgenefunction.Havingahigh-qualityreferencegenomeprovidesmanybenefitsinexploitingA.mexicanusasamodelspecies.Amongthemoreimportantuses,fromthestandpointofutilizingthesystemtounderstandevolutionarymechanisms,areinmappingandidentifyinggenesresponsibleforphenotypicchangeandinunveilinggenomicstructuralvariation(SV)thatprovidesasubstrateforadaptiveselection.Thus,todemonstratetheadvantagesAstyanaxmexicanus2.0bringstothefield,wehaveexploreditsuseineachofthesesettings,inparticularexaminingthegeneticunderpinningsofalbinismandreducedeyesize,andthenlookingatthecontributionofgenomicdeletiontovariationwithinA.mexicanuspopulations.First,withrespecttoidentificationofevolutionaryimportantgenes,wetookadvantageofacriticalattributeofA.mexicanus,nearlyuniqueamongcaveanimals:theabilitytointerbreedcavefishandsurfacefishpopulationstogeneratefertilehybridsthatcanbeintercrossedtoperformQTLstudies(forreviewseerefs.36,37).Theimprovedcontiglengthofthesurfacefishgenomeandoursyntenicanalysisthatunitesthephysicalgenometoapreviouslypublishedlinkagemapbasedongenotyping-by-sequencing(GBS)markers30shouldgreatlyaidthecorrectidentificationofgeneticchangeslinkedtocavetraits.Tothisend,wedemonstratethepoweroftheAstyanaxmexicanus2.0referenceingainingdeeperinsightintoanalreadymappedtrait(albinism)andincarryingoutnewQTLanalysisofeyereduction.AlbinismhaspreviouslybeenmappedinQTLstudiesofboththePachònandMolinopopulations20.Thisworkwascarriedoutbeforetheexistenceofanyreferencegenomeforthespecies.However,thesemappingstudiesshowedthatasinglemajorQTLineachpopulationcolocalizedwithacandidategeneoca2.Oca2loss-of-functionmutationsareknowntocausealbinisminotherorganisms,includinghumans,mice,andzebrafish,anddeletionscausingfunctionalinactivationofOca2wereidentifiedinthealbinofishfromboththeMolinoandPachóncaves20.Thesecompellingdataindicatingthatoca2mutationsarecausalofalbinismincavefishweresubsequentlyconfirmedbyCRISPR-mediatedmutagenesisinsurfacefish27.Tobuildontheseresults,wefirstperformedtwoseparatedenovoQTLanalysesofsurface/PachónF2hybrids(groupAandB)inthecontextofthenewreferencegenome(SupplementaryFig. 5a).BothstudiesidentifiedasingleQTLforalbinism,withaLODscoreof47.31atthepeakmarkeronlinkagegroup(LG)3(69%varianceexplained)ingroupA,andwithLODscoreof22.56atthepeakmarkeronLG21(37.8%varianceexplained)ingroupB(SupplementaryFig. 5b,c,f,g).AtthepeakQTLpositions,F2hybridshomozygousforthecaveallelearealbino(SupplementaryFig. 5d,h).MappingthemarkersassociatedwitheachoftheseLGstothesurfacefishgenomerevealedthattheyarebothlocatedonsurfacefishchromosome13(SupplementaryTable 4,SupplementaryFig. 5e,i).Thisdemonstratesasignificantimprovementovermappingtotheoriginalcavefishgenome:LG3fromgroupAalmostcompletelycorrespondstosurfacefishchromosome13inAstyanaxmexicanus2.0,whereasitissplitupintomanycontigsinAstyanaxmexicanus1.0.2(SupplementaryFigs. 5e,iand6a).TheQTLidentifiedonLG21ingroupBcorrespondstoa2.8 Mbregiononsurfacefishchromosome13asdeterminedusingthesequencesflankingthe1.5-LODsupportintervalasinputforEnsemblbasiclocalalignmentsearchtool(SupplementaryTable 4,SupplementaryFig. 5i).Inlinewithpreviousmappingstudies,thegeneoca2isfoundwithinthisregiononsurfacefishchromosome13.Wefurtherfunctionallyverifiedthatachangeintheoca2locusisresponsibleforalbinismintheF2hybrids,bycrossinganalbinosurface/PachónF2hybridwithageneticallyengineeredsurfacefishheterozygousforadeletioninoca2exon2127.Wefoundthat46.5%oftheoffspringcompletelylackedpigment(n = 40/86,SupplementaryFig. 5j).Thisconfirmsthatachangeintheoca2locusisresponsibleforlossofpigmentinthemappingpopulation.Whileupuntilthispointourreanalysisofalbinismwaslargelyconfirmatory,anadvantageofusingthesurfacefishgenomeasareference,comparedtothepreviouslyavailablecavefishgenome,istheabilitytomakecomparisonsbetweenregionsthataredeletedincavefishandassucharenotavailableforsequencealignment.WeutilizedtheAstyanaxmexicanus2.0referencegenometoalignsequencingdataobtainedfromwild-caughtfishfromdifferentsurfaceandcavelocalities32andanalyzedtheoca2locus(SupplementaryFigs. 7and8).WefoundthatnineoutoftenPachóncavefishcarrythedeletioninexon24thatwaspreviouslyreportedinlaboratory-raisedfish(SupplementaryFig. 720).Noneoftheindividualsfromotherpopulationsforwhichsequenceinformationwasavailablecarriedthesamedeletion.Consistentwiththelaboratorystrains,exon21ofoca2isabsentinallMolinocavefishsamples(n = 9,SupplementaryFig. 8).Noneoftheothersequencedpopulationsamplesharborthesamedeletionofexon21;however,wefoundsmallerheterozygousorhomozygousdeletionsinexon21insomewildsamplesofPachón(n = 5/9),RíoChoy(n = 1/9),andTinaja(5/9)fish(SupplementaryFig. 8).Insummary,wewereabletodetectdeletionsinoca2thatwouldhavenotbeendiscoveredwithalignmentstoAstyanaxmexicanus1.0.2sinceexon24ismissinginPachónandthusnoreference-basedalignmentswereproducedinthatregion.WeuncoveredadditionalinformationbyremappingpreviouslypublishedQTL18,19,20,21,22,23,24usingtheAstyanaxmexicanus2.0reference.Atotalof1060outof1124markers(94.3%)mappedsuccessfullywithBLASTandwereincludedinoursurfacefishQTLdatabase.Therewere77markersthatdidnotmaptothecavefishgenome28butdidmaptothesurfacefishgenome,52markersthatmappedtothecavefishbutnottheAstyanaxmexicanus2.0reference,and12markersthatdidnotmaptoeitherreference.Theimprovedcontiguityofthechromosome-levelsurfacefishassemblyallowedustoidentifyseveraladditionalcandidategenesassociatedwithQTLmarkersthatwerenotpreviouslyidentifiedinthemorefragmentedcavefishgenome(SupplementaryData 1).Forexample,themarkersAm205DandAm208Emappedtoan~1 Mbregionofchromosome6ofthesurfacefishgenome(46,516,926–47,425,713 bp)butdidnotmaptotheAstyanaxmexicanus1.0.2reference.Thisregionisassociatedwithfeedingangle21,eyesize,vibrationattractionbehavior(VAB),suborbitalneuromasts23,andmaxillarytoothnumber19.Multiplecandidategenesrelatedtocave-specificphenotypesarecontainedinthisregionincludingrhodopsin(vision),ubiad1(eyedevelopment),aswellasGABAAreceptordelta,whichisassociatedwithavarietyofbehaviorsandcouldconceivablybeinvolvedinVAB.Notably,thescaffoldcontainingthesefourgenesinAstyanaxmexicanus1.0.2(KB871939.1)wasnotlinkedtothisQTL,demonstratingtheutilityofincreasedcontiguityofAstyanaxmexicanus2.0.Previousstudiessuggestedthatthegeneretinalhomeoboxgene3(rx3)lieswithintheQTLforouterplexiformlayeroftheeye24.AnotherQTLforeyesize,sizeofthethirdsuborbitalbone,andbodycondition18,19mayalsocontainrx3(lowmarkerdensityandlowpowerofolderstudiesresultinabroadQTLcriticalregioninthisarea).TheincreasedcontiguityofAstyanaxmexicanus2.0revealedthatrx3iswithintheregionencompassedbythisQTL,whereasinAstyanaxmexicanus1.0.2,themarkerforthisQTL(Am55A)andrx3werelocatedonseparatescaffolds;thus,wecouldnotappreciatethatthisQTLandkeygeneforeyedevelopmentwereinrelativelyclosegenomicproximity.Whilenoaminoacidcodingchangesareapparentbetweencavefishandsurfacefish,expressionofrx3isreducedinPachóncavefishrelativetosurfacefish28,38.Inzebrafish,rx3isexpressedintheeyefieldoftheanteriorneuralplateduringgastrulationandhasanessentialroleforthefatespecificationbetweeneyeandtelencephalon39,40.Wehavecomparedrx3expressionattheendofgastrulationandconfirmedPachónembryoshavereducedexpressiondomainsize(Fig. 2a,b).TheexpressionareaissignificantlysmallerinPachónembryoscomparedtostage-matchedsurfacefishembryos.Theexpressionofrx3isrestoredinF1hybridsbetweencavefishandsurfacefish,indicatingarecessiveinheritanceincavefish(Fig. 2a,b).Totestforaputativeroleofrx3ineyedevelopmentinA.mexicanus,weusedCRISPR/Cas9tomutatethisgeneinsurfacefish(Fig. 2c),andassessedinjected,crispantfishforeyephenotypes.Wild-typesurfacefishhavelargeeyes(Fig. 2d).Incontrast,externallyvisibleeyesarecompletelyabsentinadultCRISPantsurfacefish(n = 5,Fig. 2d).Eyephenotypeswerealsoreducedduringdevelopment.Sixdayspostfertilizationrx3crispantfishhavesmallereyesthanfishinjectedwithCas9mRNAaloneoruninjectedsiblings(SupplementaryFig. 9).Thisisconsistentwithworkfromotherspecies,inwhichmutationsinrx3(fish)orRx(mice)resultinacompletelackofeyes41,42,43.Togetherthesedatasuggestthattheroleofrx3ineyedevelopmentisconservedinA.mexicanus.Further,theysupportthehypothesisthatregulatorychangesinthisgenemaycontributetoeyelossincavefishthroughspecificationofasmallereyefield,andsubsequently,productionofasmallereye.Fig.2:Rx3analysis.aPhotographsofinsituhybridizationsforrx3(arrows)andpax2mRNAattailbudstageonsurfacefish,Pachóncavefish,andF1hybridbetweensurfacefishmaleandPachóncavefishfemale.Scalebar:200 µm.bComparisonofexpressionareaofrx3attailbudstageonsurfacefish(meanvalue34876.3 µm2,SD ± 5767.2,n = 17),Pachón(meanvalue28674.8 µm2,SD ± 5175.0,n = 14),andF1hybrid(meanvalue32617.8 µm2,SD ± 4917.2,n = 11).SignificanceisshownforunpairedStudent’sttest(two-tailed);p = 0.004isdenotedby*andnsmeansnotsignificant(p = 0.294).cDiagramofrx3gene.Boxesindicateexonandlinesindicateintrons.TheemptyboxesareUTRandthefilledboxesarecodingsequence(genebuildfromNCBI).AgRNAwasdesignedtargetingexon2.ThegRNAtargetsiteisinblueandthePAMsiteisinred.ThearrowindicatesthepredictedCas9cutsite.dLeft:adultWTcontrol(uninjectedsibling)surfacefish.Right:adultrx3CRISPR/Cas9-injectedsurfacefishlackingeyes(CRISPant).FullsizeimageInadditiontothecompileddatabaseofolderQTLstudies,anumberofgenomicintervalsassociatedwithpreviouslydescribedlocomotoractivitydifferencebetweencavefishandsurfacefish25werealsorescreenedusingthesurface-anchoredlocationsofmarkersfromthehigh-densitylinkagemap30.WithinthisPachón/surfaceQTLmap,weconfirmedthepresenceof20previouslyreportedcandidategenes,andidentified96additionalgeneswithrelevantGOterms,includingrx3,furtherdemonstratingthepowerandutilityofthisgenomicresource(SupplementaryTable 5).Thenewcandidatesincludeadditionalopsins(opn7a,opn8a,opn8b,tmtopsa,andtwoputativegreen-sensitiveopsins),aswellasseveralgenescontributingtocircadianrhythmicity(id2b,nfil3-5,cipcb,clocka,andnpas2).Whileanalysesofexpressiondataandsequencevariationarenecessarytodeterminewhichofthesecandidatesexhibitmeaningfuldifferencesbetweenmorphs,thepresenceofclockaandnpas2intheseintervalsisofparticularnote,astheoriginalanalysisconductedusingAstyanaxmexicanus1.0.2didnotprovideanyevidenceofapotentialroleformembersofthecorecircadianclockworkinmediatingobserveddifferencesinlocomotoractivitypatternsbetweenPachónandsurfacefish15,25.Finally,togainnewinsightintotheeyereductionphenotype,weusedtheAstyanaxmexicanus2.0referencetodenovogeneticallymapeyesizeinthetwosurface/PachónF2groupsthatweusedtomapalbinism.Insurface/PachónF2mappingpopulationA(n = 188),weidentifiedmultipleQTLsfornormalizedeyeperimeterthatwerespreadacrossfourLGs(Fig. 3a).TheQTLonLG1issignificantaboveathresholdofp  5%.WefoundthattheMolinopopulationhasthefewestheterozygousdeletions,whileRíoChoysurfacefishhavetheleasthomozygousdeletions,mirroringtheheterozygosityofsinglenucleotidepolymorphisms32(SupplementaryFig. 13).Inaddition,theTinajapopulationshowedthemostindividualvariabilityofeitherallelicstate(standarddeviationof427).PachónandTinajacontainedthehighestnumberofprotein-codinggenesalteredbyadeletioninatleastonehaplotype,whileRíoChoyhadtheleast(Fig. 4a).Intwoexamples,per3andephx2,wefindthedeletionsthatpresumablyalteredprotein-codinggenefunctionvariedinpopulationrepresentation,numberofbasesaffected,andhaplotypestateforeach(Fig. 4b,c).Ofthe412genesthatcontaineddeletions,109haveassignedgeneontologyincavefish(SupplementaryTable 7).Wetestedthese109genesforcanonicalpathwayenrichmentusingWebGestalt50andfoundgenessignificantlyenriched(p 50-foldwasgeneratedbasedonanestimatedgenomesizeof1.3 Gb.AllSMRTsequencesareavailableunderNCBIBioProjectnumberPRJNA533584.AssemblyanderrorcorrectionFordenovoassemblyofallSMRTsequences,weusedafuzzydeBruijngraphalgorithm,wtdbg29,forcontiggraphconstruction,followedbycollectiverawreadalignmentforassemblybaseerrorcorrection.Allerror-pronereadswerefirstusedtogenerateanassemblygraphwithgenomick-mersuniquetoeachreadthatresultsinprimarycontigs.Assembledprimarycontigswerecorrectedforrandombaseerror,predominatelyindels,usingallrawreadsmappedwithminimap265.Tofurtherreduceconsensusassemblybaseerror,wecorrectedhomozygousinsertion,deletion,andsinglebasedifferencesusingdefaultparametersettingsinPilon66with~60×coverageofanIlluminaPCR-freelibrary(150 bpreadlength)derivedfromAsty152DNA.AssemblyscaffoldingToscaffolddenovoassembledcontigs,wegeneratedaBioNanoIrysrestrictionmapofanothersurfacefish(Asty168)thatallowedsequencecontigstobeorderedandoriented,andpotentialmisassembliestobeidentified.Asty168istheoffspringoftwosurfacefish,onefromtheRíoValles(Asty02)andtheotherfromtheRíoSabinaslocality(Asty04),bothAsty02andAsty04weretheoffspringofwild-caughtfish(SupplementaryFig. 1b).WepreparedHMW-DNAinagarplugsusingapreviouslyestablishedprotocolforsofttissues67.Briefly,wefollowedaseriesofenzymaticreactionsthat(1)lysedcells,(2)degradedproteinandRNA,and(3)addedfluorescentlabelstonickedsitesusingtheIrysPrepReagentKit.ThenickedDNAfragmentswerelabeledwithAlexaFluor546dye,andtheDNAmoleculeswerecounter-stainedwithYOYO-1dye.ThelabeledDNAfragmentswereelectrophoreticallyelongatedandsizedonasingleIrysChip,andsubsequentimaginganddataprocessingdeterminedthesizeofeachDNAfragment.Finally,aBioNanoproprietaryalgorithmperformedadenovoassemblyofalllabeledfragments > 150 kbpintoawhole-genomeopticalmapwithdefinedoverlappatterns.Theindividualmapwasclusteredandscoredforpairwisesimilarity,andEuclidiandistancematriceswerebuilt.Manualrefinementswerethenperformedaspreviouslydescribed67.ChromosomebuildsUponchimericcontigcorrectionandcompletionofscaffoldassemblystepsusingtheBioNanomap,weusedChromonomer68toalignallpossiblescaffoldstotheA.mexicanushigh-densitylinkagemap30,thenassignedchromosomecoordinates.Usingdefaultparametersettings,Chromonomerattemptstofindthebestsetofnonconflictingmarkersthatmaximizesthenumberofscaffoldsinthemap,whileminimizingorderingdiscrepancies.TheoutputisaFASTAfileformatdescribingthelocationofscaffoldsbychromosome:a“chromonome”.DefiningsyntenicregionsbetweencaveandsurfacegenomesAtotalof2235GBSmarkers30weremappedtoboththecave(Astyanaxmexicanus1.0.2)andsurfacefish(Astyanaxmexicanus2.0)assemblies(SupplementaryFig. 2).TheseGBSmarkersweremappedtoAstyanaxmexicanus1.0.2usingtheEnsembl“BLAST/BLATsearch”webtoolandresultinginformationfromeachindividualquerywastranscribedintoanExcelworksheet.Tomaptothesurfacefishgenome,theNCBI“Magic-BLAST”(version1.3.0)commandlinemappingtool69wasused.Theresultingoutputofthemappingwasasingle,tabularformattedspreadsheet,whichwasusedtovisualizesyntenicregionsfromtheconstructedlinkagemap30.WeusedCircossoftwaretovisualizethepositionsofallmarkersthatmappedtotheAstyanaxmexicanus2.0genome70.Anychromonomeerrorsdetectedthroughthesesyntenyalignmentswereinvestigatedandmanuallycorrectediforthologousdatawereavailable,suchasAstyanaxmexicanus1.0.2scaffoldalignment.Moreover,tofacilitatefutureinvestigationsintothelocationofpreviouslydiscoveredQTLineachform,weprovidethecorrespondingcoordinatesoftheassemblytoassemblyalignmentsforAstyanaxmexicanus1.0.2andAstyanaxmexicanus2.0(SupplementaryData 1).GeneannotationTheAstyanaxmexicanus2.0assemblywasannotatedusingthepreviouslydescribedNCBI71andEnsembl35pipelines,includingmaskingofrepeatspriortoabinitiogenepredictionsandRNAseqevidence-supportedgenemodelbuilding.NCBIandEnsemblgeneannotationreliedonanextensivevarietyofpubliclyRNAseqdatafrombothcaveandsurfacefishtissuestoimprovegenemodelaccuracy.TheAstyanaxmexicanus2.0RefSeqorEnsemblrelease98geneannotationreportseachprovideafullaccountingofallmethodologydeployedandtheiroutputmetricswithineachrespectivebrowser.AssayinggenomequalityusingpopulationgenomicsamplesTounderstandtheimpactreferencesequencebiasandqualitymayhaveondownstreampopulationgenomicanalysesforthecavefishandsurfacefishgenomes,weutilizedthepopulationgenomicresequencedindividualsprocessedinHermanetal.32.Inbrief,100 bpsequenceswerealignedtothereferencegenomesAstyanaxmexicanus1.0.2andAstyanaxmexicanus2.0.TheNCBIannotationpipelineofbothassembliesincludedWindowMaskerandRepeatMaskerstepstodelineateandexcluderepetitiveregionsfromgenemodelannotation.ThepositionalcoordinatesforrepeatsidentifiedbyRepeatMaskerareprovidedintheBEDformatatNCBIforeachgenome.WindowMasker’s“nmer”files(counts)wereusedtoregeneraterepetitiveregionBEDcoordinates31.TheBEDcoordinatesforbothmaskerswereintersectedwithBEDToolsv2.27.172.WeusedSAMtools1.9withthesecoordinatesandalignmentqualityscorestofilteralignmentsandgeneratesummarystatisticsforeachsamplealignedtothecaveandsurfacereferencegenomes73.SummaryplotsweregeneratedinR3.6.Geneticmappinginsurface × PachóncrossesTomapalbinismandeyesizetothenewsurfacegenomeandtestrobustnessinthismethodologyacrosslaboratories,twoindependentF2mappingpopulationsconsistingofsurface/Pachónhybridswereanalyzed.First,wescored188surface/PachónF2hybridsforalbinismandnormalizedeyeperimeterthatwereusedinapreviouslypublishedgeneticmappingstudy10.PhenotypeswereassessedusingmacroscopicimagesofentirefishandmeasurementswereobtainedusingImageJ74.Normalizedeyeperimeterwasdeterminedbydividingeyeperimeterbybodylength.Albinismwasscoredasabsenceofbodyandeyepigment.The25LGsconstructeddenovoinearlierstudiesofthispopulation10werescannedusingtheR(v.3.5.3)packageR/qtl(v.1.44-9)75usingthescanonefunctionformarkerslinkedwithalbinism(binarymodel)orlefteyeperimeterrelativetobodylength(normalmodel).Thegenome-wideLODsignificancethresholdwassetatthe95thpercentileof1000permutations.AllmarkersequenceswerealignedtoboththeAstyanaxmexicanus1.0.2andAstyanaxmexicanus2.0referencesusingBowtie(v.2.2.6insensitivemode)76.Circosplotsweregeneratedusingv.0.69-670.ThesecondF2mappingpopulation(n = 219)consistedofthreeclutchesproducedfrombreedingpairedF1surface/Pachónhybridsiblings77.Albinismandeyesizewereassessedusingmacroscopicimagesofentirefish.EyediameterandfishlengthweremeasuredinImageJ74accordingtoref.78.Fishlackinganeye(21/195ontheleftside,22/172ontherightside)werenotincludedintheanalysisofeyesizeinordertoanalyzeeyesizeusinganormaldistributionmodel.Wefoundthatfishlengthwaspositivelycorrelatedwitheyediameter.ToeliminatetheeffectoffishlengthonpotentialQTL,weanalyzedeyediameternormalizedtostandardlength.WeusedR/qtl75toscantheLGs(scanonefunction)formarkerslinkedwithalbinism(binarymodel)oreyediameterrelativetobodylength(normalmodel)andassessedstatisticalsignificanceoftheLODscoresbycalculatingthe95thpercentileofgenome-widemaximumpenalizedLODscoreusing1000randompermutations.WeestimatedconfidenceintervalsfortheQTLusing1.5-LODsupportinterval(Iodintfunction).Complementationanalysisinalbinosurface–PachónF2fishAnalbinosurface–PachónF2fishwascrossedtooca2Δ4bp/+surfacefish27.Fivedaypostfertilization,larvaloffspringfromthiscrosswerescoredforpigmentation(presenceorabsenceofmelaninpigmentation)byroutineobservation.Larvaewereimagedunderadissectingmicroscopeandthenumberofpigmentedandalbinoprogenywereassessed.Followingthis,DNAwasextractedfromeightpigmentedandeightalbinoprogeny,andthesefishwerethengenotypedfortheengineered4 bpdeletionbyPCRfollowedbygelelectrophoresis,usinglocusspecificprimers(SupplementaryTable 8—ComplementationPrimer)andmethods79,80.Briefly,larvalfishwereeuthanizedinMS-222,fishwereimaged,thenDNAwasextractedfromwholefish,PCRwasperformedfollowedbygelelectrophoresis.Allgenotypedalbinoembryoshadtwobands(indicatingthepresenceoftheengineereddeletion),whereasallgenotypedpigmentedembryoshadasingleband,indicatinginheritanceofthewild-typeallelefromthesurfacefishparent.GeneticmappingofpreviouslymappedQTLstudiestothesurfacefishgenomeToidentifyanycandidategenespotentiallyassociatedwithcavephenotypesthatwerenotevidentinthemorefragmentedAstyanaxmexicanus1.0.2genome,wegeneratedaQTLdatabasefortheAstyanaxmexicanus2.0genometoidentifygenomicregionscontaininggroupsofmarkersassociatedwithcave-derivedphenotypes.Wefollowedthemethods32usedtocreateasimilardatabaseforthecavefishgenome.BLASTwasusedtoidentifylocationsinthesurfacefishgenomefor1156markersfromseveralpreviousQTLstudies18,19,21,22,23,81.ThetopBLASThitforeachmarkerwasidentifiedbyrankinge-value,followedbybitscore,andthenalignmentlength.Previously,687ofthesemarkersweremappedtothecavefishgenome.BEDToolsintersectandthesurfacefishgenomeannotationwasusedtoidentifyallgeneswithintheregionsofinterest.GenomicintervalsassociatedwithactivityQTLpreviouslyidentifiedusingtheexistinghigh-densitylinkagemapweresimilarlyreexaminedusingtheestablishedlocationsofthe2235GBSmarkerswithintheAstyanaxmexicanus2.0genome25.WetheninvestigatedwhethergenesidentifiednearQTLhadontologiesassociatedwithcavephenotypesusingtheNCBI(https://www.ncbi.nlm.nih.gov/)orEnsemblgenomebrowser(https://www.ensembl.org).However,geneswerereferredtoascandidatesbecausetheyarewithintheQTLconfidenceinterval.Evenifagenewithintheintervaldoesnothavepreviouslyknownfunctionscontributingtothetraitofinterest,itdoesnoteliminateitasapotentialcandidate.Generationofrx3CRISPantfishCRISPantsurfacefishweregeneratedusingCRISPR/Cas9.AgRNAtargetingexon2wasdesignedandgeneratedasdescribedpreviously82.Briefly,oligoAcontainingthegRNAsequence(5′-GTGTAGCTGAAACGTGGTGA-3′)betweenthesequencefortheT7promoterandasequenceoverlappingwiththesecondoligo,oligoB,wassynthesized(IDT)(SupplementaryTable 8—CRISPR-Oligo).FollowingannealingwithOligoBandamplification,theT7MegascriptKit(Ambion)wasusedtotranscribethegRNAwithseveralmodifications,asinrefs.27,80.Specifically,themanufacturer’sdirectionswerefollowedexceptthatthefollowingreagentswereaddedpertranscriptionreaction:10 μlofRNAse-freewater,5 μlofDNAtemplate,1 μlofeachnucleosidetriphosphate,1 μlof10xtranscriptionbuffer,and1 μlofT7polymeraseenzymemix.ThegRNAwascleanedupusingthemiRNeasyminikit(Qiagen)followingmanufacturer’sinstructionsandelutedintoRNase-freewater.Nls-Cas9-nls83mRNAwastranscribedusingthemMessagemMACHINET3kit(LifeTechnologies)followingmanufacturer’sdirections.Single-cellembryoswereinjectedwith25 pgofgRNAand150 pgCas9mRNA.Injectionswereperformedusingacapillaryglassneedlemountedinamicromanipulatorconnectedtoamicroinjector.Injectionpressurewasadjustedwitheveryneedletoachievea~1.0 nlinjectionvolume79,80.InjectedfishwerescreenedtoassessformutagenesisbyPCRusingprimerssurroundingthegRNAtargetsite(SupplementaryTable 8—CRISPantprimer).GenotypingwasperformedwithDNAextractedfromwholelarvalfishfromtheclutch,asdescribedpreviously84.Briefly,DNAwasisolatedbyincubatingsamplesin50 mMNaOHat95 °Cfor30 min,afterwhicha1/10thvolumeofTris-HCLpH8wasadded.FollowingPCR,gelelectrophoresiswasusedtodiscriminatebetweenalleleswithindels(morethanonebandresultsinasmearyband)andwild-typealleles(singleband:asinref.79).Foradultfish,injectedfish(crispants)alongwithuninjectedwild-typesiblingswereraisedtoadulthood.Eyesizewasassessedinbothrx3gRNA/Cas9mRNAinjectedfish(crispants)andwild-typesiblingadultfish(n = 5each).FishwereanesthetizedinMS-222andimagedunderadissectingmicroscope.Forlarvalfish,eyesizewasassessedatday6postfertilizationandcomparedbetweenrx3CRISPantfish,150 pgCas9mRNAonlyinjectedsiblings,andwild-typeuninjectedsiblings(n = 10each)usingFiji74andcorrectedforstandardlength.EyeareawascomparedbetweengroupsusinganANOVAfollowedbyposthocTukeytests.StatisticswereperformedinGraphPadPrism.WholemountinsituhybridizationandimagingFulllengthofrx3codingsequencewasamplifiedbyPCRfromsurfacefishcDNA(SupplementaryTable 8—WMISH-primer).TheantisenseprobewasgeneratedusingadigoxigeninRNALabelingKit(RocheDiagnostics,Indianapolis,IN).Insituhybridizationwasperformedwith65%formamideofthehybridizationmix,andtheincubationtemperaturewassetat65 °C.Thestainedembryosweredehydratedthroughasuccessiveseriesofmethanolandstoredat−20 °C.Beforeimaging,theembryosweregraduallyrehydratedintoPBST(PBSwith0.1%Tween-20).Duringtheimaging,theembryosweresuspendedin0.2%agarinPBS,andtheexpressionareaofrx3wasmeasuredusingNISElementsF2.30softwarewithastereomicroscope(SMZ1500,Nikon).SVdetectionWefirstalignedIlluminareadsfromA.mexicanuspopulationresequencingdatadescribedbyHermanetal.32totheAstyanaxmexicanus2.0(SRAaccession:PRJNA260715)referenceusingBWA-MEMv0.7.1785withdefaultoptions.Wethenconvertedtheoutputtobamformat,fixedmatepairinformation,sortedandremovedduplicatesusingtheSAMtoolsv1.9view-bh,fixmate-m,sort,andmarkdup-rmodules,respectively73.Werantwosoftwarepackagesforcallingstructuralvariantsusingthesealignments:manta48andlumpy49.Weranmantav1.6.0withdefaultoptionsassuggestedinthepackage’smanual,andlumpyusingthesmoovev0.2.3pipeline(https://github.com/brentp/smoove)usingthecommandsgiveninthe“Populationcalling”section.Nextflow86workflowsforourrunningofbothSVcallerscanbefoundathttps://github.com/esrice/workflows.Duetothelowsequencecoverage(average~9×)availablepersample,weconfinedanalysistodeletionscalledbybothSVcallersandwithinthesizerange500–100 kb.Weconsideradeletiontobepresentinbothsetsofcallsiftherewasareciprocaloverlapof50%ofthelengthofthedeletion.Weusedthescriptsmerge_deletions.pytofindtheintersectionofthetwosetsofdeletions,annotate_vcf.pytogroupthedeletionsbytheireffect(i.e.,deletionofcoding,intronic,regulatory,orintergenicsequence),andcount_variants_per_sample.pytocountthenumbersofvariantscalledineachsample;allscriptsarepublicly available(seeCodeAvailability).Allprotein-codinggeneswithdetecteddeletionsamongcavefishpopulationsasdefinedinthisstudywereusedasinputtotestforsignificantenrichmentamongspecificdatabaseswithinWebGestalt50.EntrezgeneIDswereinputasgenesymbols,withorganismofinterestsettozebrafishusingprotein-codinggenesasthereferenceset.PathwaycommondatabasesofKEGGandPanthercanonicalgenesignalingpathwaysaswellasthevariousgenesassociatedwithdiseasescuratedinOMIMandDigenet87werereportedusingahypergeometrictest,andthesignificancelevelwassetat0.05.WeimplementedtheBenjaminiandHochbergmultipletestadjustment88tocontrolforfalsediscovery.ReportingsummaryFurtherinformationonresearchdesignisavailableinthe NatureResearchReportingSummarylinkedtothisarticle. Dataavailability TherawsequencingandfinalassemblydatageneratedinthisstudyhavebeendepositedintheNCBIBioProjectdatabaseunderaccessioncodePRJNA89115.TheresequencingdatausedinthisstudyareavailableintheNCBIBioProjectdatabaseunderaccessioncodePRJNA260715.OriginaldataunderlyingthismanuscriptcanbeaccessedfromtheStowersOriginalDataRepositoryathttp://www.stowers.org/research/publications/libpb-1528,foundwithinthearticleormadeavailablebytheauthorsonrequest. Codeavailability AllcodesusedtogeneratethedatainthispaperareavailableonGitHub:https://doi.org/10.5281/zenodo.4433170. References1.Culver,D.&Pipan,T.TheBiologyofCavesandOtherSubterraneanHabitats2ndedn,(OxfordUniversityPress,2019).2.Protas,M.&Jeffery,W.R.Evolutionanddevelopmentincaveanimals:fromfishtocrustaceans.WileyInterdiscip.Rev.Dev.Biol.1,823–845(2012).PubMed  PubMedCentral  Article  GoogleScholar  3.Emerling,C.A.&Springer,M.S.Eyesunderground:regressionofvisualproteinnetworksinsubterraneanmammals.Mol.Phylogenet.Evol.78,260–270(2014).CAS  PubMed  Article  GoogleScholar  4.Yoshizawa,M.,McGaugh,S.E.&Keene,A.BiologyandEvolutionoftheMexicanCavefish(AcademicPress,2015).5.Gross,J.B.ThecomplexoriginofAstyanaxcavefish.BMCEvol.Biol.12,105(2012).PubMed  PubMedCentral  Article  GoogleScholar  6.Maldonado,E.,Rangel-Huerta,E.,Rodriguez-Salazar,E.,Pereida-Jaramillo,E.&Martinez-Torres,A.Subterraneanlife:behavior,metabolic,andsomeotheradaptationsofAstyanaxcavefish.J.Exp.Zool.BMol.Dev.Evol.https://doi.org/10.1002/jez.b.22948(2020).7.Krishnan,J.&Rohner,N.Cavefishandthebasisforeyeloss.Philos.Trans.R.Soc.Lond.BBiol.Sci.372,https://doi.org/10.1098/rstb.2015.0487(2017).8.Aspiras,A.C.,Rohner,N.,Martineau,B.,Borowsky,R.L.&Tabin,C.J.Melanocortin4receptormutationscontributetotheadaptationofcavefishtonutrient-poorconditions.Proc.NatlAcad.Sci.USA112,9668–9673(2015).ADS  CAS  PubMed  Article  GoogleScholar  9.Jaggard,J.B.etal.HypocretinunderliestheevolutionofsleeplossintheMexicancavefish.Elife7,https://doi.org/10.7554/eLife.32637(2018).10.Stockdale,W.T.etal.HeartregenerationintheMexicancavefish.CellRep.25,1997–2007e1997(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  11.Riddle,M.R.etal.Insulinresistanceincavefishasanadaptationtoanutrient-limitedenvironment.Nature555,647–651(2018).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  12.Yoshizawa,M.etal.Theevolutionofaseriesofbehavioraltraitsisassociatedwithautism-riskgenesincavefish.BMCEvol.Biol.18,89(2018).PubMed  PubMedCentral  Article  CAS  GoogleScholar  13.Elipot,Y.,Hinaux,H.,Callebert,J.&Retaux,S.Evolutionaryshiftfromfightingtoforaginginblindcavefishthroughchangesintheserotoninnetwork.Curr.Biol.23,1–10(2013).CAS  PubMed  Article  GoogleScholar  14.Gross,J.B.&Powers,A.K.Anaturalanimalmodelsystemofcraniofacialanomalies:theblindMexicancavefish.Anat.Rec.303,24–29(2020).Article  GoogleScholar  15.Carlson,B.M.&Gross,J.B.CharacterizationandcomparisonofactivityprofilesexhibitedbythecaveandsurfacemorphotypesoftheblindMexicantetra,Astyanaxmexicanus.Comp.Biochem.Physiol.CToxicol.Pharm.208,114–129(2018).CAS  Article  GoogleScholar  16.Xiong,S.,Krishnan,J.,Peuss,R.&Rohner,N.EarlyadipogenesiscontributestoexcessfataccumulationincavepopulationsofAstyanaxmexicanus.Dev.Biol.441,297–304(2018).CAS  PubMed  Article  GoogleScholar  17.Rohner,N.Cavefishasanevolutionarymutantmodelsystemforhumandisease.Dev.Biol.441,355–357(2018).CAS  PubMed  Article  GoogleScholar  18.Protas,M.etal.Multi-traitevolutioninacavefish,Astyanaxmexicanus.Evol.Dev.10,196–209(2008).PubMed  Article  GoogleScholar  19.Protas,M.,Conrad,M.,Gross,J.B.,Tabin,C.&Borowsky,R.RegressiveevolutionintheMexicancavetetra,Astyanaxmexicanus.Curr.Biol.17,452–454(2007).CAS  PubMed  PubMedCentral  Article  GoogleScholar  20.Protas,M.E.etal.Geneticanalysisofcavefishrevealsmolecularconvergenceintheevolutionofalbinism.Nat.Genet.38,107–111(2006).CAS  PubMed  Article  GoogleScholar  21.Kowalko,J.E.etal.ConvergenceinfeedingpostureoccursthroughdifferentgeneticlociinindependentlyevolvedcavepopulationsofAstyanaxmexicanus.Proc.NatlAcad.Sci.USA110,16933–16938(2013).ADS  CAS  PubMed  Article  GoogleScholar  22.Kowalko,J.E.etal.Lossofschoolingbehaviorincavefishthroughsight-dependentandsight-independentmechanisms.Curr.Biol.23,1874–1883(2013).CAS  PubMed  PubMedCentral  Article  GoogleScholar  23.Yoshizawa,M.,Yamamoto,Y.,O’Quin,K.E.&Jeffery,W.R.Evolutionofanadaptivebehavioranditssensoryreceptorspromoteseyeregressioninblindcavefish.BMCBiol.10,108(2012).PubMed  PubMedCentral  Article  GoogleScholar  24.O’Quin,K.E.,Yoshizawa,M.,Doshi,P.&Jeffery,W.R.QuantitativegeneticanalysisofretinaldegenerationintheblindcavefishAstyanaxmexicanus.PLoSONE8,e57281(2013).ADS  PubMed  PubMedCentral  Article  CAS  GoogleScholar  25.Carlson,B.M.,Klingler,I.B.,Meyer,B.J.&Gross,J.B.GeneticanalysisrevealscandidategenesforactivityQTLintheblindMexicantetra,Astyanaxmexicanus.PeerJ6,e5189(2018).PubMed  PubMedCentral  Article  CAS  GoogleScholar  26.Stahl,B.A.etal.StabletransgenesisinAstyanaxmexicanususingtheTol2transposasesystem.Dev.Dyn.https://doi.org/10.1002/dvdy.32(2019).27.Klaassen,H.,Wang,Y.,Adamski,K.,Rohner,N.&Kowalko,J.E.CRISPRmutagenesisconfirmstheroleofoca2inmelaninpigmentationinAstyanaxmexicanus.Dev.Biol.441,313–318(2018).CAS  PubMed  Article  GoogleScholar  28.McGaugh,S.E.etal.Thecavefishgenomerevealscandidategenesforeyeloss.Nat.Commun.5,5307(2014).ADS  CAS  PubMed  PubMedCentral  Article  GoogleScholar  29.Ruan,J.&Li,H.Fastandaccuratelong-readassemblywithwtdbg2.Nat.Methods17,155–158(2020).CAS  PubMed  Article  GoogleScholar  30.Carlson,B.M.,Onusko,S.W.&Gross,J.B.Ahigh-densitylinkagemapforAstyanaxmexicanususinggenotyping-by-sequencingtechnology.G35,241–251(2014).PubMed  Article  GoogleScholar  31.Morgulis,A.,Gertz,E.M.,Schaffer,A.A.&Agarwala,R.WindowMasker:window-basedmaskerforsequencedgenomes.Bioinformatics22,134–141(2006).CAS  PubMed  Article  GoogleScholar  32.Herman,A.etal.Theroleofgeneflowinrapidandrepeatedevolutionofcave-relatedtraitsinMexicantetra,Astyanaxmexicanus.Mol.Ecol.27,4397–4416(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  33.Koren,S.etal.Denovoassemblyofhaplotype-resolvedgenomeswithtriobinning.Nat.Biotechnol.https://doi.org/10.1038/nbt.4277(2018).34.Sayers,E.W.etal.DatabaseresourcesoftheNationalCenterforBiotechnologyInformation.NucleicAcidsRes.https://doi.org/10.1093/nar/gkz899(2019).35.Yates,A.D.etal.Ensembl2020.NucleicAcidsRes.https://doi.org/10.1093/nar/gkz966(2019).36.O’QuinK,M.S.ThegeneticbasesoftroglomorphyinAstyanax:howfarwehavecomeandwheredowegofromhere?inBiologyandEvolutionoftheMexicanCavefish(edsKeene,A.,Yoshizawa,M.&McGaugh,S.E.)(Elsevier,2015).37.Casane,D.&Retaux,S.EvolutionarygeneticsofthecavefishAstyanaxmexicanus.Adv.Genet.95,117–159(2016).CAS  PubMed  Article  GoogleScholar  38.Sifuentes-Romero,I.etal.RepeatedevolutionofeyelossinMexicancavefish:evidenceofsimilardevelopmentalmechanismsinindependentlyevolvedpopulations.J.Exp.Zool.BMol.Dev.Evol.https://doi.org/10.1002/jez.b.22977(2020).39.Cavodeassi,F.,Ivanovitch,K.&Wilson,S.W.Eph/Ephrinsignallingmaintainseyefieldsegregationfromadjacentneuralplateterritoriesduringforebrainmorphogenesis.Development140,4193–4202(2013).CAS  PubMed  PubMedCentral  Article  GoogleScholar  40.Stigloher,C.etal.Segregationoftelencephalicandeye-fieldidentitiesinsidethezebrafishforebrainterritoryiscontrolledbyRx3.Development133,2925–2935(2006).CAS  PubMed  Article  GoogleScholar  41.Mathers,P.H.,Grinberg,A.,Mahon,K.A.&Jamrich,M.TheRxhomeoboxgeneisessentialforvertebrateeyedevelopment.Nature387,603–607(1997).ADS  CAS  PubMed  Article  GoogleScholar  42.Loosli,F.etal.Medakaeyelessisthekeyfactorlinkingretinaldeterminationandeyegrowth.Development128,4035–4044(2001).CAS  PubMed  GoogleScholar  43.Loosli,F.etal.Lossofeyesinzebrafishcausedbymutationofchokh/rx3.EMBORep.4,894–899(2003).CAS  PubMed  PubMedCentral  Article  GoogleScholar  44.Martinez-Morales,J.R.etal.ojoplano-mediatedbasalconstrictionisessentialforopticcupmorphogenesis.Development136,2165–2175(2009).CAS  PubMed  Article  GoogleScholar  45.Borowsky,R.Restoringsightinblindcavefish.Curr.Biol.18,R23–R24(2008).CAS  PubMed  Article  GoogleScholar  46.Yang,C.H.etal.NEAP/DUSP26suppressesreceptortyrosinekinasesandregulatesneuronaldevelopmentinzebrafish.Sci.Rep.7,5241(2017).ADS  PubMed  PubMedCentral  Article  CAS  GoogleScholar  47.Stahl,B.A.&Gross,J.B.AcomparativetranscriptomicanalysisofdevelopmentintwoAstyanaxcavefishpopulations.J.Exp.Zool.BMol.Dev.Evol.328,515–532(2017).CAS  PubMed  PubMedCentral  Article  GoogleScholar  48.Chen,X.etal.Manta:rapiddetectionofstructuralvariantsandindelsforgermlineandcancersequencingapplications.Bioinformatics32,1220–1222(2016).CAS  PubMed  Article  GoogleScholar  49.Layer,R.M.,Chiang,C.,Quinlan,A.R.&Hall,I.M.LUMPY:aprobabilisticframeworkforstructuralvariantdiscovery.GenomeBiol.15,R84(2014).PubMed  PubMedCentral  Article  GoogleScholar  50.Wang,J.,Vasaikar,S.,Shi,Z.,Greer,M.&Zhang,B.WebGestalt2017:amorecomprehensive,powerful,flexibleandinteractivegenesetenrichmentanalysistoolkit.NucleicAcidsRes.45,W130–W137(2017).CAS  PubMed  PubMedCentral  Article  GoogleScholar  51.Pinero,J.etal.TheDisGeNETknowledgeplatformfordiseasegenomics:2019update.NucleicAcidsRes.https://doi.org/10.1093/nar/gkz1021(2019).Article  PubMedCentral  PubMed  GoogleScholar  52.Prober,D.A.,Rihel,J.,Onah,A.A.,Sung,R.J.&Schier,A.F.Hypocretin/orexinoverexpressioninducesaninsomnia-likephenotypeinzebrafish.J.Neurosci.26,13400–13410(2006).CAS  PubMed  PubMedCentral  Article  GoogleScholar  53.Yokogawa,T.etal.Characterizationofsleepinzebrafishandinsomniainhypocretinreceptormutants.PLoSBiol.5,e277(2007).PubMed  PubMedCentral  Article  CAS  GoogleScholar  54.Lin,L.etal.Thesleepdisordercaninenarcolepsyiscausedbyamutationinthehypocretin(orexin)receptor2gene.Cell98,365–376(1999).CAS  PubMed  Article  GoogleScholar  55.Chen,C.W.,Lin,J.&Chu,Y.W.iStable:off-the-shelfpredictorintegrationforpredictingproteinstabilitychanges.BMCBioinform.14,S5(2013). GoogleScholar  56.Cheng,J.,Randall,A.&Baldi,P.Predictionofproteinstabilitychangesforsingle-sitemutationsusingsupportvectormachines.Proteins62,1125–1132(2006).CAS  PubMed  Article  GoogleScholar  57.Waterhouse,A.etal.SWISS-MODEL:homologymodellingofproteinstructuresandcomplexes.NucleicAcidsRes.46,W296–W303(2018).CAS  PubMed  PubMedCentral  Article  GoogleScholar  58.Suno,R.etal.Crystalstructuresofhumanorexin2receptorboundtothesubtype-selectiveantagonistEMPA.Structure26,7–19e15(2018).CAS  PubMed  Article  GoogleScholar  59.Schartl,M.etal.Thegenomeoftheplatyfish,Xiphophorusmaculatus,providesinsightsintoevolutionaryadaptationandseveralcomplextraits.Nat.Genet.45,567–572(2013).CAS  PubMed  PubMedCentral  Article  GoogleScholar  60.Chain,F.J.etal.Extensivecopy-numbervariationofyounggenesacrosssticklebackpopulations.PLoSGenet.10,e1004830(2014).PubMed  PubMedCentral  Article  GoogleScholar  61.Catanach,A.etal.ThegenomicpoolofstandingstructuralvariationoutnumberssinglenucleotidepolymorphismbythreefoldinthemarineteleostChrysophrysauratus.Mol.Ecol.28,1210–1223(2019).CAS  PubMed  Article  GoogleScholar  62.Flagel,L.E.,Willis,J.H.&Vision,T.J.ThestandingpoolofgenomicstructuralvariationinanaturalpopulationofMimulusguttatus.GenomeBiol.Evol.6,53–64(2014).PubMed  Article  GoogleScholar  63.Noda,M.etal.RoleofPer3,acircadianclockgene,inembryonicdevelopmentofmousecerebralcortex.Sci.Rep.9,5874(2019).ADS  PubMed  PubMedCentral  Article  CAS  GoogleScholar  64.Li,S.B.&deLecea,L.Thehypocretin(orexin)system:fromaneuralcircuitryperspective.Neuropharmacology167,107993(2020).CAS  PubMed  Article  GoogleScholar  65.Li,H.Minimapandminiasm:fastmappinganddenovoassemblyfornoisylongsequences.Bioinformatics32,2103–2110(2016).CAS  PubMed  PubMedCentral  Article  GoogleScholar  66.Walker,B.J.etal.Pilon:anintegratedtoolforcomprehensivemicrobialvariantdetectionandgenomeassemblyimprovement.PLoSONE9,e112963(2014).ADS  PubMed  PubMedCentral  Article  CAS  GoogleScholar  67.Lam,E.T.etal.Genomemappingonnanochannelarraysforstructuralvariationanalysisandsequenceassembly.Nat.Biotechnol.30,771–776(2012).CAS  PubMed  Article  GoogleScholar  68.Catchen,J.,Amores,A.&Bassham,S.Chromonomer:atoolsetforrepairingandenhancingassembledgenomesthroughintegrationofgeneticmapsandconservedsynteny.https://doi.org/10.1101/2020.02.04.934711(2020).69.Boratyn,G.M.,Thierry-Mieg,J.,Thierry-Mieg,D.,Busby,B.&Madden,T.L.Magic-BLAST,anaccurateRNA-seqalignerforlongandshortreads.BMCBioinform.20,405(2019).Article  CAS  GoogleScholar  70.Krzywinski,M.etal.Circos:aninformationaestheticforcomparativegenomics.GenomeRes.19,1639–1645(2009).CAS  PubMed  PubMedCentral  Article  GoogleScholar  71.Pruitt,K.D.etal.RefSeq:anupdateonmammalianreferencesequences.NucleicAcidsRes.42,D756–D763(2014).CAS  PubMed  Article  GoogleScholar  72.Quinlan,A.R.&Hall,I.M.BEDTools:aflexiblesuiteofutilitiesforcomparinggenomicfeatures.Bioinformatics26,841–842(2010).CAS  PubMed  PubMedCentral  Article  GoogleScholar  73.Li,H.etal.Thesequencealignment/mapformatandSAMtools.Bioinformatics25,2078–2079(2009).PubMed  PubMedCentral  Article  CAS  GoogleScholar  74.Schindelin,J.etal.Fiji:anopen-sourceplatformforbiological-imageanalysis.Nat.Methods9,676–682(2012).CAS  Article  GoogleScholar  75.Broman,K.W.,Wu,H.,Sen,S.&Churchill,G.A.R/qtl:QTLmappinginexperimentalcrosses.Bioinformatics19,889–890(2003).CAS  PubMed  Article  GoogleScholar  76.Langmead,B.&Salzberg,S.L.Fastgapped-readalignmentwithBowtie2.Nat.Methods9,357–359(2012).CAS  PubMed  PubMedCentral  Article  GoogleScholar  77.Riddle,M.R.etal.GeneticarchitectureunderlyingchangesincarotenoidaccumulationduringtheevolutionoftheblindMexicancavefish,Astyanaxmexicanus.J.Exp.Zool.BMol.Dev.Evol.https://doi.org/10.1002/jez.b.22954(2020).78.Hinaux,H.etal.AdevelopmentalstagingtableforAstyanaxmexicanussurfacefishandPachoncavefish.Zebrafish8,155–165(2011).PubMed  Article  GoogleScholar  79.Kowalko,J.E.,Ma,L.&Jeffery,W.R.GenomeeditinginAstyanaxmexicanususingtranscriptionactivator-likeeffectornucleases(TALENs).J.Vis.Exp.https://doi.org/10.3791/54113(2016).80.Stahl,B.A.etal.ManipulationofgenefunctioninMexicancavefish.J.Vis.Exp.https://doi.org/10.3791/59093(2019).81.Yoshizawa,M.etal.Distinctgeneticarchitectureunderliestheemergenceofsleeplossandprey-seekingbehaviorintheMexicancavefish.BMCBiol.13,15(2015).PubMed  PubMedCentral  Article  GoogleScholar  82.Varshney,G.K.etal.High-throughputgenetargetingandphenotypinginzebrafishusingCRISPR/Cas9.GenomeRes.25,1030–1042(2015).CAS  PubMed  PubMedCentral  Article  GoogleScholar  83.Jao,L.E.,Wente,S.R.&Chen,W.EfficientmultiplexbialleliczebrafishgenomeeditingusingaCRISPRnucleasesystem.Proc.NatlAcad.Sci.USA110,13904–13909(2013).ADS  CAS  PubMed  Article  GoogleScholar  84.Ma,L.,Jeffery,W.R.,Essner,J.J.&Kowalko,J.E.GenomeeditingusingTALENsinblindMexicancavefish,Astyanaxmexicanus.PLoSONE10,e0119370(2015).PubMed  PubMedCentral  Article  CAS  GoogleScholar  85.Li,H.Aligningsequencereads,clonesequencesandassemblycontigswithBWA-MEM.https://arxiv.org/abs/1303.3997(2013).86.DiTommaso,P.etal.Nextflowenablesreproduciblecomputationalworkflows.Nat.Biotechnol.35,316–319(2017).Article  CAS  GoogleScholar  87.Pinero,J.etal.DisGeNET:acomprehensiveplatformintegratinginformationonhumandisease-associatedgenesandvariants.NucleicAcidsRes.45,D833–D839(2017).CAS  PubMed  Article  GoogleScholar  88.Benjamini,Y.Controllingthefalsediscoveryrate:apracticalandpowerfulapproachtomultipetesting.J.R.Stat.Soc.Ser.BStat.Methodol.57,289–300(1995).MathSciNet  MATH  GoogleScholar  89.Humphrey,W.,Dalke,A.&Schulten,K.VMD:visualmoleculardynamics.J.Mol.Graph.14,33–38(1996).CAS  PubMed  Article  GoogleScholar  DownloadreferencesAcknowledgementsTheauthorswouldliketothankKarinZueckert-GaudenzandMihaelaSardiufortechnicalassistanceandtheStowersaquaticsgroupforfishcareandhelpwithshipmentsoffishbetweenNYUandStowers.A.K.,S.E.M.,andN.R.aresupportedbyNIH1R01GM127872-01.ThisworkwasalsosupportedbyNIHR24OD011198toW.C.W.,L.H.,E.S.R.,T.G.-L.,andM.K.,NSFEDGEAward1923372toN.R.,J.E.K.,andS.E.M.,NSFDEB-1754231toJ.E.K.andA.K.,andNSFIOS-1933428toJ.E.K.,S.E.M.,andN.R.AuthorN.R.isfurthersupportedbyinstitutionalfunding,fundingfromtheEdwardMallinckrodtFoundation,andNIHDP2AG071466.J.E.K.isfurthersupportedbyNIHR15HD099022.J.B.G.issupportedbyNIDCRR01-DE025033andNSFDEB-1457630.Somecomputationforthisworkwasperformedonthehigh-performancecomputinginfrastructureprovidedbyResearchComputingSupportServicesandinpartbytheNationalScienceFoundationundergrantnumberCNS-1429294attheUniversityofMissouri,Columbia,MO,USA.TheMinnesotaSupercomputingInstituteattheUniversityofMinnesotaprovidedresourcesthatcontributedtotheresearchresultsreportedwithinthispaper.AuthorinformationAuthornotesRobertPeußPresentaddress:InstituteforEvolutionandBiodiversity,UniversityofMünster,Münster,GermanyMistyR.RiddlePresentaddress:DepartmentofBiology,UniversityofNevada,Reno,NV,USAAffiliationsDepartmentofAnimalSciences,InstituteforDataScienceandInformatics,BondLifeSciencesCenter,UniversityofMissouri,Columbia,MO,USAWesleyC.WarrenDepartmentofSurgery,InstituteforDataScienceandInformatics,BondLifeSciencesCenter,UniversityofMissouri,Columbia,MO,USAWesleyC.WarrenDepartmentofBiologicalSciences,UniversityofCincinnati,Cincinnati,OH,USATylerE.Boggs & JoshuaB.GrossDepartmentofBiology,NewYorkUniversity,NewYork,NY,USARichardBorowskyDepartmentofBiologicalSciences,NorthernKentuckyUniversity,HighlandHeights,KY,USABrianM.CarlsonHarrietL.WilkesHonorsCollege,FloridaAtlanticUniversity,Jupiter,FL,USAEstephanyFerrufino, JohannaE.Kowalko, ItzelSifuentes-Romero, BethanyA.Stanhope & SunishkaThakurDepartmentofGenomeSciences,UniversityofWashington,Seattle,WA,USALaDeanaHillierDepartmentofPhysiology,AnatomyandGenetics,UniversityofOxford,Oxford,UKZhilianHu & MathildaT.M.MommersteegDepartmentofBiologicalSciences,FloridaAtlanticUniversity,Jupiter,FL,USAAlexC.Keene & BethanyA.StanhopeStowersInstituteforMedicalResearch,KansasCity,MO,USAAlexanderKenzior, RobertPeuß & NicolasRohnerMcDonnellGenomeInstitute,WashingtonUniversity,StLouis,MO,USAChadTomlinson, MilinnKremitzki & TinaGraves-LindsayBioinfo,Plantagenet,ON,CanadaMadeleineE.LemieuxDepartmentofEcology,Evolution,andBehavior,UniversityofMinnesota,SaintPaul,MN,USASuzanneE.McGaugh, JeffreyT.Miller & RachelL.MoranBondLifeSciencesCenter,UniversityofMissouri,Columbia,MO,USAEdwardS.RiceGeneticsDepartment,BlavatnikInstitute,HarvardMedicalSchool,Boston,MA,USAMistyR.Riddle & CliffordJ.TabinDepartmentofCellandDevelopmentalBiology,UniversityCollegeLondon,London,UKYoshiyukiYamamotoDepartmentofMolecular&IntegrativePhysiology,KUMedicalCenter,KansasCity,KS,USANicolasRohnerAuthorsWesleyC.WarrenViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarTylerE.BoggsViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarRichardBorowskyViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarBrianM.CarlsonViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarEstephanyFerrufinoViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarJoshuaB.GrossViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarLaDeanaHillierViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarZhilianHuViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarAlexC.KeeneViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarAlexanderKenziorViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarJohannaE.KowalkoViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarChadTomlinsonViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarMilinnKremitzkiViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarMadeleineE.LemieuxViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarTinaGraves-LindsayViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarSuzanneE.McGaughViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarJeffreyT.MillerViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarMathildaT.M.MommersteegViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarRachelL.MoranViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarRobertPeußViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarEdwardS.RiceViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarMistyR.RiddleViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarItzelSifuentes-RomeroViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarBethanyA.StanhopeViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarCliffordJ.TabinViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarSunishkaThakurViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarYoshiyukiYamamotoViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarNicolasRohnerViewauthorpublicationsYoucanalsosearchforthisauthorin PubMed GoogleScholarContributionsW.C.W.andN.R.conceivedofthestudy.Allauthorsperformedandanalyzedtheexperiments.W.C.W.,N.R.,R.B.,B.M.C.,J.B.G.,A.C.K.,J.E.K.,S.E.M.,M.T.M.M.,R.P.,M.R.R.,C.T.,andY.Y.wrotethepaper.CorrespondingauthorsCorrespondenceto WesleyC.WarrenorNicolasRohner.Ethicsdeclarations Competinginterests Theauthorsdeclarenocompetinginterests. AdditionalinformationPeerreviewinformationNatureCommunicationsthanksJeramiahSmith,DeshouWang,andMasatoYoshizawafortheircontributiontothepeerreviewofthiswork.Peerreviewreportsareavailable.Publisher’snoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations.Supplementaryinformation SupplementaryInformationPeerReviewFileDescriptionsofAdditionalSupplementaryFilesSupplementaryData1ReportingSummaryRightsandpermissions OpenAccessThisarticleislicensedunderaCreativeCommonsAttribution4.0InternationalLicense,whichpermitsuse,sharing,adaptation,distributionandreproductioninanymediumorformat,aslongasyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.Theimagesorotherthirdpartymaterialinthisarticleareincludedinthearticle’sCreativeCommonslicense,unlessindicatedotherwiseinacreditlinetothematerial.Ifmaterialisnotincludedinthearticle’sCreativeCommonslicenseandyourintendeduseisnotpermittedbystatutoryregulationorexceedsthepermitteduse,youwillneedtoobtainpermissiondirectlyfromthecopyrightholder.Toviewacopyofthislicense,visithttp://creativecommons.org/licenses/by/4.0/. ReprintsandPermissionsAboutthisarticleCitethisarticleWarren,W.C.,Boggs,T.E.,Borowsky,R.etal.Achromosome-levelgenomeofAstyanaxmexicanussurfacefishforcomparingpopulation-specificgeneticdifferencescontributingtotraitevolution. NatCommun12,1447(2021).https://doi.org/10.1038/s41467-021-21733-zDownloadcitationReceived:23June2020Accepted:02February2021Published:04March2021DOI:https://doi.org/10.1038/s41467-021-21733-zSharethisarticleAnyoneyousharethefollowinglinkwithwillbeabletoreadthiscontent:GetshareablelinkSorry,ashareablelinkisnotcurrentlyavailableforthisarticle.Copytoclipboard ProvidedbytheSpringerNatureSharedItcontent-sharinginitiative Furtherreading GeneticmappingofmetabolictraitsintheblindMexicancavefishrevealssex-dependentquantitativetraitlociassociatedwithcaveadaptation MistyR.Riddle ArielAspiras CliffordJ.Tabin BMCEcologyandEvolution(2021) Fishgenomicsanditsimpactonfundamentalandappliedresearchofvertebratebiology SyedFarhanAhmad MaryamJehangir CesarMartins ReviewsinFishBiologyandFisheries(2021) CommentsBysubmittingacommentyouagreetoabidebyourTermsandCommunityGuidelines.Ifyoufindsomethingabusiveorthatdoesnotcomplywithourtermsorguidelinespleaseflagitasinappropriate. DownloadPDF Advertisement Explorecontent Researcharticles Reviews&Analysis News&Comment Videos Collections Subjects FollowusonFacebook FollowusonTwitter Signupforalerts RSSfeed Aboutthejournal Aims&Scope Editors JournalInformation Openaccess Articleprocessingcharges EditorialValuesStatement JournalImpact Editors'Highlights Contact Top50Articles Editorialpolicies Publishwithus Forauthors ForReviewers Submitmanuscript Search Searcharticlesbysubject,keywordorauthor Showresultsfrom Alljournals Thisjournal Search Advancedsearch Quicklinks Explorearticlesbysubject Findajob Guidetoauthors Editorialpolicies