Single-cone imaging in inherited and acquired colour vision deficiencies

Single-cone imaging in inherited and acquired colour vision deficiencies

Rigmor C Baraas, Hilde R Pedersen and Lene A Hagen

Colourvisiondeficienciesarecommoninhumansandoccur bothasaconsequenceofinheritedconeopsinmutations, alteringthenumberorfunctionofthedifferentconetypes expressedintheretina,andacquiredthroughsecondary disruptionofconefunctionandstructure.Thisreviewdescribes recentadvancesmadeinunderstandingcolourvision deficienciesfromcombiningknowledgeaboutconeopsin geneswithsingle-coneimaginginlivinghumans.Examination oftheeffectoftheopsingenemutationsupontheconemosaic andcolourvisionphenotypesshowsthatnotallinheritedcolour visiondeficienciesarestationaryandsomeinheritedcongenital eyediseasesmaycauseimpairedcolourvisionasa

consequenceofarresteddevelopment.

Address

NationalCentreforOptics,VisionandEyeCare,FacultyofHealthand SocialSciences,UniversityofSouth-EasternNorway,Hasbergsvei36, 3616Kongsberg,Norway

Correspondingauthor:Baraas,RigmorC(rigmor.baraas@usn.no)

CurrentOpinioninBehavioralSciences2019,30:55–59 ThisreviewcomesfromathemedissueonVisualperception EditedbyHannahSmithsonandJohnSWerner

https://doi.org/10.1016/j.cobeha.2019.05.006

2352-1546/ã2019TheAuthors.PublishedbyElsevierLtd.Thisisan openaccessarticleundertheCCBYlicense(http://creativecommons.

org/licenses/by/4.0/).

Introduction

Directvisualisationofsingle photoreceptorcellsin eyes of living humans was achieved for the first time in 1999[1],pavingthewayforabroaderanddeeperunder- standing of the cone mosaic and colour vision. In-vivo microscopic imaging of the retina was made possible through adding adaptive optics (AO) — a wavefront sensortomeasuretheocularmonochromaticaberrations andadeformablemirrortocorrectfortheaberrations—in ophthalmoscopicimagingsystems.Severalsystemshave been developedovertheyears,butitisthemultimodal versionofAOscanninglightophthalmoscope(AOSLO), incorporating both confocaland non-confocal detection thathashelpedusmakethemostimportantdiscoveries [2,3].Such asystemallowsfor simultaneousimagingof both theconfocalwaveguided lightfromtheouter seg- ments and non-confocal backscattered light from the

retina, revealing the structure of the inner segmentsof thecones(indirecttemporalandspatialcorrespondence asillustratedinFigure1).Thisisofsignificance,because remnant innersegmentscannow beobservedin condi- tionswhere coneopsinmutations rendertheouter seg- ments dysfunctionalor non-functionalso thatit cannot waveguide lightproperly. Single-coneimagingin living human eyes has allowed for major advancements in understanding the properties of the cone mosaic, the distribution and organisation of different cone types [4],structural and functionalchanges as aconsequence ofdisruptionscausedbygeneticmutationsordisease.It has become a tool that allows for tracking longitudinal changes of the cone mosaic, including the effect of medicationand genetictreatment [5].Examination of the effect of the opsin gene mutations upon the cone mosaic and colour vision phenotypesshows thatnot all inherited colour vision deficiencies are stationary or benign [6–8,9] and some inherited eye diseases may causeimpairedcolourvisionasaconsequenceofarrested development[10,11].Inthispaper,wediscussrecent advancementsinunderstandingthelinksbetweenretinal structure, function and colour perception, elucidating associated consequences of colour vision loss, either as aresultofopsingenemutationsorsecondarytodiseaseor geneticmutationsthatdisruptretinaldevelopment.

Studying the variation in cone photoreceptor mosaic in living humans with impaired colour vision

Therearethreelinesofworkinthestudyofvariationsin theconephotoreceptormosaicandcolourvisiondeficien- cies that we will focus on. Firstly, we will review the differencesbetweenconeopsinmutationsthatgiverise to inherited colour vision deficiencies (impaired colour discrimination,butnotcolourblindness)andtheireffect on retinal structure and natural history (stationary or progressive). The aim here is to understand why some inherited red–green (protan/deutan) and blue–yellow (tritan) colour vision deficiencies may be progressive.

Wewillthenbrieflytouchuponvariationinconephoto- receptormosaicininheritedconedysfunctionsthatgive rise tocolourblindness (achromatopsia),and theimpor- tanceofindividualassessmentwhenconsideringwhowill possibly benefit fromgenetic treatment to restore cone function. Finally, we will mention the effect of other genetic mutations that arrest the development of the central partsof theretina, and thestructure–perception correlations between single-cone imaging, optical

coherencetomography(OCT)imagingandcolourvision impairment.

The cone mosaic in inherited red–green colour vision deficiencies

Inheritedcolourvisiondeficienciesareaconsequenceof cone opsin mutations. Mutations in the genes for the humanlong-wavelength(L) orthe middle-wavelength (M)coneopsin,localisedontheX-chromosomeatXq28 (genetic designation OPN1LW and OPN1MW), give risetored–greencolourvisiondeficiencies.Mutationsin the genes for the human short-wavelength (S) cone opsin, localised to an autosome on chromosome 7 at 7q32(geneticdesignationOPN1SW),giverisetoblue–

yellow colourvision deficiencies [12].There is a large variationinthearrangementofconeopsingenesgiving rise to large differences in the associated degree of colourvisiondeficiency.Two typesofopsinmutations havebeenassociatedwithred–greendichromaticcolour vision; one encodes partial or complete deletion or replacement of photopigment of a different spectral class,the otherencodes a photopigment that does not function normally (random missense mutation).

Recently, several lines ofwork have been carried out to elucidate how these mutations might be expressed

and what effect they might have on the cone mosaic.

Carroll et al. used AO flood-illuminated imaging to examine the cone mosaic in red–green dichromatic individuals [7,8]. The participants harboured either a mutation that encodes the replacement of L with M photopigment, but with a deleterious combination of nucleotides at normal polymorphic positions (LIAVA, oftenreferredtoasanL/Minterchangemutation)[12], or the mutation C203R¹ in the M opsin gene, that encodes a non-functional photopigment [7,8]. They showed that the cone density associated with the LIAVAmutation wassimilar tothat innormaltrichro- mats,albeitwithamottledandlessregularappearance.

The C203R mutation was associated with lower cone density,butaregularconemosaicinterleavedwithdark regionsthoughttohavebeendegeneratedconesexpres- sing the non-functional photopigment. The LIAVA conesappearedtobestillpresent,havingperhapsslowly progressively lost function, whereas the C203R cones had completely degenerated early in foveal develop- ment. Further studies with confocalAOSLO imaging, includinganotherrandommissensemutation(W177R),

Figure1

25 µm

0 1 2 3 4 5N

1 T

(a)

(b)

(c)

(d)

200 µm

IS

OS

Current Opinion in Behavioral Sciences

Single-coneimagingoflivinghumans.(a)OCTimageshowingthelayeredstructureofthecentralretinafrom1.5temporally(T)to5nasally(N) withthecharacteristicfovealpit(centredat0)ofapersonwithnormalspatialandcolourvision.Scalebaris200mm.Thelayersincludingthe inner(blue)andouter(red)segmentstructuresareoutlined.(b)Asketchofaconephotoreceptorwithitsinnerandoutersegments.Acone photoreceptorisabout50mmlong,anditsinnersegmentisabout1mmindiameterinthefoveaandabout2mmat5eccentricityinahealthy adultretina.AOSLOimagesofnon-confocalbackscatteredlightfromtheretina,revealingthestructureof(c)theinnersegmentsandtheconfocal waveguidedlightfrom(d)theoutersegmentsofthecones.Scalebaris25mm.

1Missensemutationwherethecysteineresidueataminoacidposition 203ofthepigmentmoleculeisreplacedwitharginine.

deletionofexonsandotherL/Minterchangemutations (LVAVA, LIAVS, LVVVA), have expanded on these conclusions [13,14]. L/M interchange mutations and exon2deletionswereshowntobeassociatedwithmore disruptive retinal structure changesin the foveacom- pared withrandom missensemutations. This included damage to neighbouring S cones and rods. Recently, multimodal AOSLO imaging of participants with X-linked cone opsin mutations, associated with red–

greencolourvisiondeficiency(bothanomaloustrichro- macy and dichromacy) [15], has shown that the dark regions observed in confocal images appears to be regionswith reduced orlostcone function (poor orno waveguiding of light), because intact cone inner seg- mentsareclearlyvisible[9](seeFigure2b).Furtherto this, the large between-individual variation in retinal structure, functionandcolourperceptionappearstobe associated withboth the typesof L/M opsinmutation and the ratio of expression of first versus downstream genes onthegene array [15].

The cone mosaic in inherited blue–yellow colour vision deficiencies

Thetypesofopsinmutationsassociatedwithblue–yellow colourvisiondeficiencyappearmoresimilartomutations in the geneencoding rhodopsin (therod photoreceptor pigment)givingrisetoretinitispigmentosa[6,16],thanto LandMopsinmutations,occurringataminoacidposi- tions expected to result in photoreceptor degeneration [12]. Baraas et al. used AO flood-illuminated imaging including retinal densitometry to examine the cone mosaic in a family with an S opsin mutation [6]. The eldestmemberwasdiagnosedtobeadichromatlacking Sconefunction(tritanopia),whereasayoungermember onlyhadamilddegreeoftritancolourvisiondeficiency [6]. Images of their photoreceptor mosaic showed that cone density associated with this specific mutation (R283Q) was within normallimits, but the eldest trita- nopic member of the family lackedS conesand had a more irregular mosaic than observed in the younger familymemberandinnormalcontrols.TheR283Qcones appear tobeprogressivelydegenerating,withtritancol- our vision deficiencyonly beingmanifest whenasuffi- cient number of S cones have degenerated. A similar observation was made with another S opsin mutation (T190I)inthatolderfamilymembersweremoreseverely affectedthanthosewhowereyounger[16].Resultsfrom assessing colourvision functionatdifferentlightlevels, beforeandafterbleachinganddarkadaptation,corrobo- ratethesuggestionthatSopsinmutationsaremoreakin

Figure2

25 µm

Normal

Aniridia Achromatopsia Tritan (T190I)

Protanope (LIAVA)

200 μm

0 1 2 3 4 5

1

(a)

(b)

(c)

(d)

(e)

T N N N

N

T T T

T

N

0 1 2 3 4 5

1

0 1

1 3 2 4 5

0 1 2 3 4 5

1

Retinal location (deg)

0 1

2 1 3 4 5

25 μm

Current Opinion in Behavioral Sciences

Single-coneimagesoflivinghumanswithdifferentdegreesofspatial andcolourvision.TheleftcolumnshowsOCTimagesrevealingthe layeredstructureofthecentralretinaofpersonswith(a)normalcolour visionandnormalconeopsingenes,(b)red–green(protan)colour visiondeficiencycausedbyanMinterchangemutation(LIAVA, subjectMM0142[9]),(c)blue–yellow(tritan)colourvisiondeficiency causedbyanSopsinmutation(T190I,subjectK1539[16]),(d) achromatopsia(colourblind)and(e)aniridia(subject5120[11]).

Scalebaris200mm.Thedisruptioninthelayerwiththeouter segmentsisclearlyvisibleinachromatopsia(d)asisthefoveal hypoplasia(lackoffovealpit)inaniridia(e).Thecorrespondingsingle- coneimagesofeachparticipantareshowninthemiddlecolumn, revealingthestructureoftheinner(top)andoutersegments(bottom) ofconeandrodphotoreceptorsatabout5eccentricity(markedby thewhitearrows).Scalebaris25mm.Theimagesshowthe differencesinobservedconedensitywhenconeopsinmutations

rendertheoutersegmentsdysfunctional(b:LIAVA)orgenetic mutationsrendertheoutersegmentsnon-functional(d:

achromatopsia)notwaveguidinglightproperly.Theimagesintheright columnarealterationsofaphotographtakenbyBaraasofEthelScull 36Times(1963)byAndyWarholtosimulatethecorrespondinglossof spatialandcolourvision.

to rhodopsin mutations, and that there is a progressive element.Figure2cshowimagesoftheretinalstructurein a participant with the T190I S opsin mutation. Cone density appears normal, but retinal densitometry is needed to ascertainwhether some of thedark areason the confocal image are poorly functional S cone outer segments.Notallinheritedtritancolourvisiondeficien- ciesmaybehaveinthesameway;itwillbedowntothe type of opsin mutation and whateffectthis has onthe cone photoreceptor function at a given time in life.

Longitudinal imaging with multimodal AOSLO, com- binedwithretinaldensitometry,arerequiredtoimprove understanding to what degree cone inner segment structureisretained inSconedysfunction.

Theadvancementsbroughtforwardbycombininganal- yses of opsin genotype with colour vision and photore- ceptor phenotypes from single-cone imaging in living humans have been imperative for understanding the potentialforgenetherapy andto restoreconefunction.

Although some may deemthis not to beimportant for thosewithastationaryandbenignprotan,deutanortritan colourvisiondeficiency,thereisastrongcaseforunder- standingthewholeaspectfromvariationswithinnormal colour vision and cone mosaics through life, and even moresoforbeingabletooffergenetictreatmenttothose whoaredefactocolourblindasaconsequenceofinher- itedconedysfunction.

The cone mosaic in the colour blind (achromats/monochromats)

Achromatopsiaandblueconemonochromacy(complete andincompletecolourblindness)aretypicallystationary cone dysfunctions which give rise to impaired central visual acuity, nystagmus and photophobia. The cone mosaic, when imaged with AO flood systems [17] or confocal AOSLO [18], appears to have reduced reflec- tanceanddarkareaswhereresidualconestructuresmay reside.Withconfocalandnon-confocaldetectionAOSLO systems,ithasbeenshownthatthereareintactconeinner segment structures [3] in these dark areas, even when conefunctionismarkedlyreducedorlost[19],butwith considerablevariationbetweenparticipants[5],seealso Figure2d.Thedegreeofvariationintheremainsofcone structureinachromatopsia,withinthesameandbetween differentmutations,underscorestheimportanceofindi- vidualassessmentwhenconsideringrestoringconefunc- tionwithgenetherapy.

The cone mosaic and colour vision associated with arrested foveal development

Geneticmutations associatedwithalbinismand aniridia are associated with arrested development resulting in fovealhypoplasia.Thisunderdevelopmentofthecentral parts of the retina is typically associated with reduced visual acuity, nystagmus and photophobia. It has been shownthat there is acorrelationbetween cone density

andthedegreeoffovealhypoplasia,assessedbyspectral domain OCT images, in both albinism and aniridia [11,20].Thedegreeoffovealhypoplasiacorrelateswith the degree of red–green colour vision in aniridia[10], thosewithpoorestred–greensensitivityhavethelowest conedensity[11],seeFigure2e.Itisnotknownifthisis alsothecaseinalbinism.Itisassumedthatdifferencesin retinalganglion celldensity and/or cone-midget retinal ganglioncellpathwaysarefactorsthatmaycontributeto variation in colour vision between individuals with the samegradeoffoveal hypoplasia.

Conclusions

The utilisation of multimodal adaptive optics scanning light ophthalmoscopy to image single cone outer and innersegmentshasallowedustoanswersomeimportant questionsabouttheeffectsdifferenttypesofopsinmuta- tionsassociatedwithinheritedcolourvisiondeficiencies andotherinheritedeye diseaseshaveonconestructure andfunction(Figure2).Becauseofdevelopmentalmech- anisms, including epigenetic factors and experience- dependent change and individuation [21] that decide the number of cones any individual express on their retina,both cross-sectionalandlongitudinalstudies uti- lizingmultimodalsingle-coneimagingwillcontinuetobe importanttoolstogainanevenbetterunderstandingfor cellularretinalanatomyandcolourperception.Additional advancements,suchasfunctionalmeasurementsofsingle cones during imaging, are expected to transform our understandingof thecorrelations betweentheorganisa- tionofdifferentconetypesandcolourvisioninthefuture [22].Theimportanceofunderstandingthesecorrelations isnotonlyforthebenefitofthosewiththediseaseorfor treatmentofthose,butalsoforunderstandingwhatcon- stitutesnormalhealthydevelopmentandwhomightbeat riskfor developingage-relateddisease. Thereisalarge variationinthenumberofconesinthemacularregionof thosewithnormalcolourvision[23]andassociatedvaria- tionincolourvisionlosswithage[24].Thisindicatesthat thereisarealchancethatthosewithlowredundancyof conesaretheoneswho experienceearly(colour) vision loss, whereas those with high redundancy of macular cones may retain full visual function for longer. Only longitudinalstudiestrackingbothcolourvisionandmul- timodalsingle-coneimaginginlivinghumanswithknown opsingenearraywill allowustolearnmore.

Conflict of interest statement

Acknowledgements

TheauthorsthankStuartJGilson,EliseDeesKrekling,MaureenNeitzand AlfredoDubrafortheircontributiontothiswork.TheyalsothankJoseph CarrollandEmilyPattersonforsharingimagesofoneoftheirparticipants withtheLIAVAmutation.

TheauthorshavereceivedsupportfromtheResearchCouncilofNorway, theOslofjordFundandAniridiaNorwayfortheirresearch.Theresearch

facilitiesattheNationalCentreforOptics,VisionandEyeCarewere constructedandfundedthroughastrategiccommitmentbytheUniversity ofSouth-EasternNorwayincollaborationwiththeeyecareindustryin Norway.

References and recommended reading

Papersofparticularinterest,publishedwithintheperiodofreview, havebeenhighlightedas:

ofspecialinterest ofoutstandinginterest

1. RoordaA,WilliamsDR:Thearrangementofthethreecone classesinthelivinghumaneye.Nature1999,397:520-522.

2. DubraA,SulaiY:Reflectiveafocalbroadbandadaptiveoptics scanningophthalmoscope.BiomedOptExpress2011,2:1757-1768.

3. ScolesD,SulaiYN,LangloCS,FishmanGA,CurcioCA,CarrollJ, DubraA:Invivoimagingofhumanconephotoreceptorinner segments.InvestOphthalmolVisSci2014,55:4244-4251.

4. HoferH,CarrollJ,NeitzJ,NeitzM,WilliamsDR:Organizationof thehumantrichromaticconemosaic.JNeurosci2005, 25:9669-9679.

5. GeorgiouM,LittsKM,KalitzeosA,LangloCS,KaneT,SinghN, KassilianM,HirjiN,KumaranN,DubraA,CarrollJetal.:Adaptive opticsretinalimagingincnga3-associatedachromatopsia:

retinalcharacterization,interocularsymmetry,andintrafamilial variability.InvestOphthalmolVisSci2019,60:383-396.

Georgiouetal.usedmultimodalAOSLOretinalimagingtoinvestigatethe cone mosaic in participants with CNGA3-assosiated achromatopsia.

Non-confocalAOSLOimagesrevealedirregularanddisruptedremnant conemosaicwithlargevariabilityamongindividuals,evenforthesame mutations.Thepaper showstheusefulness ofmultimodalAOSLOin understandingdiseasepathogenesisinachromatopsia.Largebetween- individualvariabilityunderlinestheimportanceofassessingpersonswith ACHMonanindividualbasisinselectionforclinicaltrials.

6. BaraasRC,CarrollJ,GuntherKL,ChungM,WilliamsDR, FosterDH,NeitzM:Adaptiveopticsretinalimagingrevealss- conedystrophyintritancolor-visiondeficiency.JOptSocAm AOptImageSciVis2007,24:1438-1447.

7. CarrollJ,BaraasRC,Wagner-SchumanM,RhaJ,SiebeCA, SloanC,TaitDM,ThompsonS,MorganJI,NeitzJetal.:Cone photoreceptormosaicdisruptionassociatedwithcys203arg mutationinthem-coneopsin.ProcNatlAcadSciUSA2009, 106:20948-20953.

8. CarrollJ,NeitzM,HoferH,NeitzJ,WilliamsDR:Functional photoreceptorlossrevealedwithadaptiveoptics:analternate causeofcolorblindness.ProcNatlAcadSciUSA2004, 101:8461-8466.

9.

PattersonEJ,KalitzeosA,KasilianM,GardnerJC,NeitzJ, HardcastleAJ,NeitzM,CarrollJ,MichaelidesM:Residualcone structureinpatientswithx-linkedconeopsinmutations.Invest OphthalmolVisSci2018,59:4238-4248.

Pattersonetal.usedmultimodalAOSLOretinalimagingtoassessresi- dual conestructureinparticipantswithOPN1LWorOPN1MWopsin mutationswhichintroducesplicingdefectsleadingtoalackoforreduced amountoffunctionalphotopigmentinconesexpressingthemutation.

Confocalimagesrevealeddisruptedconereflectivityandreducedcone density, whereas non-confocal images revealed residual cone inner segmentscouldbevisualized.Thestudyunderlinestheimportanceof employingmultimodalimagingtoadvancetheknowledgeaboutretinal cellularphenotypesinX-linkedconeopsinmutations.

10. PedersenHR,HagenLA,LandsendECS,GilsonSJ,UtheimOA, UtheimTP,NeitzM,BaraasRC:Colorvisioninaniridia.Invest OphthalmolVisSci2018,59:2142-2152.

Pedersenetal.examinedtheassociationbetweencolorvisionandretinal structure in persons with congenital aniridia revealing the tight link betweencentralretinaldevelopmentandvisualfunction.

11. PedersenHR,NeitzM,GilsonSJ,LandsendECS,UtheimOA, UtheimTP,BaraasRC:Theconephotoreceptormosaicin aniridia:within-familyphenotype-genotypediscordance.

OphthalmolRetina2019,3:523-534.

Pedersenetal. usedmultimodal AOSLOas well asOCT imagingto assessvariability inconenumerosity andfovealhypoplasia structure in severalmembers ofthesame family who hadaniridia causedby deletioninthePAX6.Theresultsshowedconsiderablevariabilityincone densitywithineachgradeoffovealhypoplasiaandunderlinestheimpor- tanceofadvancingknowledgeaboutretinalcellularphenotypeinaniridia.

12. NeitzJ,NeitzM:Thegeneticsofnormalanddefectivecolor vision.VisRes2011,51:633-651.

13. CarrollJ,DubraA,GardnerJC,Mizrahi-MeissonnierL,CooperRF, DubisAM,NordgrenR,GeneadM,Connor TBJr,StepienKE etal.:Theeffectofconeopsinmutationsonretinalstructure andtheintegrityofthephotoreceptormosaic.Invest OphthalmolVisSci2012,53:8006-8015.

14. Wagner-SchumanM,NeitzJ,RhaJ,WilliamsDR,NeitzM, CarrollJ:Color-deficientconemosaicsassociatedwithxq28 opsinmutations:astopcodonversusgenedeletions.VisRes 2010,50:2396-2402.

15. PattersonEJ,WilkM,LangloCS,KasilianM,RingM,HufnagelRB, DubisAM,TeeJJ,KalitzeosA,GardnerJCetal.:Cone photoreceptorstructureinpatientswithx-linkedcone dysfunctionandred-greencolorvisiondeficiency.Invest OphthalmolVisSci2016,57:3853-3863.

Pattersonetal.usedmultimodalAOSLOimagingtoassessresidualcone structureinparticipants withOPN1LWor OPN1MWopsin mutations associatedwithconedysfunctionandcongenitalred-greencolourvision deficiencies to elucidate the genotype-phenotype link. The results revealedadirectlinkbetweenconemosaicdisruptionandOPN1LW/

MWopsinmutationsandunderlinestheimportanceofadvancingknowl- edgeaboutretinalcellularphenotypeinconedysfunctions.

16. BaraasRC,HagenLA,DeesEW,NeitzM:Substitutionof isoleucineforthreonineatposition190ofs-opsincausess- cone-functionabnormalities.VisRes2012,73:1-9.

17. DubisAM,CooperRF,AboshihaJ,LangloCS,SundaramV,LiuB, CollisonF,FishmanGA,MooreAT,WebsterARetal.:Genotype- dependentvariabilityinresidualconestructurein

achromatopsia:towarddevelopingmetricsforassessing conehealth.InvestOphthalmolVisSci2014,55:7303-7311.

18. CideciyanAV,HufnagelRB,CarrollJ,SumarokaA,LuoX, SchwartzSB,DubraA,LandM,MichaelidesM,GardnerJCetal.:

Humanconevisualpigmentdeletionssparesufficient photoreceptorstowarrantgenetherapy.HumGeneTher2013, 24:993-1006.

19. LangloCS,PattersonEJ,HigginsBP,SummerfeltP,RazeenMM, ErkerLR,ParkerM,CollisonFT,FishmanGA,KayCNetal.:

Residualfovealconestructureincngb3-associated achromatopsia.InvestOphthalmolVisSci2016,57:3984-3995.

20. LeeDJ,WoertzEN,VisotckyA,WilkMA,HeitkotterH, LindermanRE,TarimaS,SummersCG,BrooksBP,BrilliantMH etal.:Thehenlefiberlayerinalbinism:comparisontonormal andrelationshiptoouternuclearlayerthicknessandfoveal conedensity.InvestOphthalmolVisSci2018,59:5336-5348.

21. SweattJD:Theepigeneticbasisofindividuality.CurrOpin BehavSci2019,25:51-56.

22. SabesanR,HoferH,RoordaA:Characterizingthehumancone photoreceptormosaicviadynamicphotopigment

densitometry.PLoSOne2015,10:e0144891.

23. DeesEW,DubraA,BaraasRC:Variabilityinparafovealcone mosaicinnormaltrichromaticindividuals.BiomedOptExpress 2011,2:1351-1358.

24. DeesEW,GilsonSJ,NeitzM,BaraasRC:Theinfluenceofl-opsin genepolymorphismsandneuralageingonspatio-chromatic contrastsensitivityin20-71yearolds.VisRes2015,116:13-24.