Distribution of DArT, AFLP, and SSR markers in agenetic linkage map of a doubled-haploidhexaploid wheat population

Kassa Semagn, Åsmund Bjørnstad, Helge Skinnes, Anne Guri Marøy,Yalew Tarkegne, and Manilal William

Abstract: A genetic linkage mapping study was conducted in 93 doubled-haploid lines derived from a cross betweenTriticum aestivum L. em. Thell ‘Arina’ and a Norwegian spring wheat breeding line, NK93604, using diversity arraystechnology (DArT), amplified fragment length polymorphism (AFLP), and simple sequence repeat (SSR) markers. Theobjective of this study was to understand the distribution, redundancy, and segregation distortion of DArT markers incomparison with AFLP and SSR markers. The map contains a total of 624 markers with 189 DArTs, 165 AFLPs and270 SSRs, and spans 2595.5 cM. All 3 marker types showed significant (p < 0.01) segregation distortion, but it washigher for AFLPs (24.2%) and SSRs (22.6%) than for DArTs (13.8%). The overall segregation distortion was 20.4%.DArTs showed the highest frequency of clustering (27.0%) at < 0.5 cM intervals between consecutive markers, whichis 3 and 15 times higher than SSRs (8.9%) and AFLPs (1.8%), respectively. This high proportion of clustering ofDArT markers may be indicative of gene-rich regions and (or) the result of inclusion of redundant clones in thegenomic representations, which was supported by the presence of very high correlation coefficients (r > 0.98) andmulticollinearity among the clustered markers. The present study is the first to compare the utility of DArT with AFLPand SSR markers, and the present map has been successfully used to identify novel QTLs for resistance to Fusariumhead blight and powdery mildew and for anther extrusion, leaf segment incubation, and latency.

Key words: ‘Arina’, diversity arrays technology, double haploid, genetic map, marker clustering, microsatellite.

Résumé : Les auteurs ont réalisé une cartographie génétique sur 93 lignées haploïdes doublées dérivées d’un croise-ment entre le blé d’automne Triticum aestivum L. em. Thell ‘Arina’ et NK93604, une lignée norvégienne de blé deprintemps, au moyen de marqueurs DArT (« diversity arrays technology »), AFLP (« amplified fragment length poly-morphism ») et microsatellites (SSRs). L’objectif de ce travail était d’étudier la distribution, la redondance et la distor-sion de la ségrégation des marqueurs DArT par rapport aux AFLP et SSR. La carte totalisait 624 marqueurs dont189 DArT, 165 AFLP et 270 SSR et s’étendait sur 2 595,5 cM. Les 3 types de marqueurs souffraient de distorsion(à p < 0,01), mais celle-ci était plus importante chez les AFLP (24,2 %) et les SSR (22,6 %) que chez les marqueursDArT (13,8 %). Globalement, 20,4 % des marqueurs montraient une distorsion de leur ségrégation. Les marqueursDArT présentaient la plus forte incidence de groupement de marqueurs (27,0 %), avec moins de 0,5 cM entre mar-queurs voisins, ce qui est 3 à 15 fois plus que pour les marqueurs SSR (8,9 %) ou AFLP (1,8 %). Cette forte propor-tion de marqueurs groupés pourrait découler de régions riches en gènes ou l’inclusion de clones redondants au sein desreprésentations génomiques. Cette dernière hypothèse était supportée par des coefficients de corrélation très élevés (r >0,98) et une colinéarité multiple parmi les marqueurs groupés. La présente étude est la première à comparer l’utilitédes marqueurs DArT par rapport aux AFLP et SSR. De plus, la présente carte a été employée avec succès pour identi-fier de nouveaux QTL pour la résistance à la fusariose de l’épi et à l’oïdium ainsi que des QTL pour l’extrusion desanthères, la durée d’incubation et la latence des segments foliaires.

Mots clés : ‘Arina’, « diversity arrays technology », haploïde doublé, carte génétique, groupement de marqueurs,microsatellite.

[Traduit par la Rédaction] Semagn et al. 555

Genome 49: 545–555 (2006) doi:10.1139/G06-002 © 2006 NRC Canada

545

Received 20 September 2005. Accepted 13 December 2005. Published on the NRC Research Press Web site at http://genome.nrc.caon 8 June 2006.

K. Semagn, Å. Bjørnstad1, H. Skinnes, A.G. Marøy, and Y. Tarkegne. Department of Plant and Environmental Sciences,Norwegian University of Life Sciences, P.O. Box 5003, N-1432, Ås, Norway.M. William. Centro Internacional de Mejoramiento de Maizy Trigo (CIMMYT), Apdo Postal 6-641, 06600 Mexico, D.F. Mexico.

1Corresponding author (e-mail: asmund.bjornstad@umb.no).

Introduction

Hexaploid wheat (Triticum aestivum L. em. Thell) is oneof the world’s most important food crops, supplying nearly55% of the carbohydrates consumed worldwide (Gupta et al.1999). It is an allohexaploid (2n = 6x = 42) with 3 distinctbut genetically related (homoeologous) genomes, A, B, andD, each with 7 chromosomes. Wheat has a large genome,16 million kilobases per haploid cell (Bennett and Smith1976), with a <2.7% gene-containing fraction and >80% re-petitive DNA.

The advent of DNA marker technology in the 1980s of-fered a large number of environmentally insensitive geneticmarkers that could be generated to follow the inheritance ofimportant agronomic traits (Gupta et al. 1999; Peleman andvan der Voort 2003). Restriction fragment length polymor-phisms (RFLPs) (Botstein et al. 1980), random amplifiedpolymorphic DNA (RAPD) (Williams et al. 1990), micro-satellite or simple sequence repeats (SSRs) (Akkaya et al.1992), sequence-tagged-sites (STS) (Talbert et al. 1994), andamplified fragment length polymorphisms (AFLPs) (Vos etal. 1995) are the most commonly used marker types forwheat map construction. Microsatellite markers remain astandard for map construction, as they are highly polymor-phic even between closely related lines, require a smallamount of DNA, can be easily automated, allow high-throughput screening, can be exchanged between laborato-ries, and are highly transferable between populations (Guptaet al. 1999). AFLP offers a high level of utility for variouspurposes and has been used to generate large numbers ofmarkers for the construction of high-density genetic maps(e.g., Barrett and Kidwell 1998; Parker et al. 1998; Huang etal. 2000; Schwarz et al. 2000; Chalmers et al. 2001). Diver-sity arrays technology (DArT) is a microarray hybridization-based technique that enables the simultaneous typing ofseveral hundred polymorphic loci spread over the genomewithout the need of prior sequence information (Jaccoud etal. 2001; Wenzl et al. 2004). DArT has recently been used ingenetic mapping and fingerprinting studies in Arabidopsis(Wittenberg et al. 2005), barley (Wenzl et al. 2004), rice(Jaccoud et al. 2001), cassava (Xia et al. 2005), and wheat(Akbari et al. 2006). DArT is a high-throughput, quick, andhighly reproducible method.

Several genetic maps have been reported for all hexaploidwheat chromosomes (Liu and Tsunewaki 1991; Gale et al.1995; Roder et al. 1998a; Li et al. 1999; Messmer et al.1999; Mingeot and Jacquemin 1999; Gupta et al. 2002;Paillard et al. 2003; Somers et al. 2004; Song et al. 2005).The International Triticeae Mapping Initiative (ITMI) popu-lation (Synthetic × ‘Opata’; T. aestivum) is one of the mostpolymorphic wheat-mapping populations and has been ex-tensively mapped with RFLP, AFLP, and SSR markers (Nel-son et al. 1995a, 1995b, 1995c; Van Deynze et al. 1995;Marino et al. 1996; Roder et al. 1998a; Song et al. 2005). Toraise the density of microsatellite markers available on awheat genetic map, Somers et al. (2004) constructed a high-density microsatellite consensus map by mapping commonmicrosatellites on each chromosome in 4 different mappingpopulations. DArT has the potential for increasing markerdensity within a short time and at low cost (Jaccoud et al.2001; Wenzl et al. 2004). The cost of DArT per data point

has been reported to be 10-fold lower than the cost of SSR(Xia et al. 2005). However, the distribution of loci over thegenome has not been compared with other marker technolo-gies. The objective of this study was, therefore, to constructa genetic linkage map in a doubled-haploid (DH) populationusing DArT, AFLP and SSR markers to understand the dis-tribution, redundancy, and segregation distortion of DArTmarkers in comparison with AFLP and SSR markers.

Materials and methods

Plant materials and DNA extractionA DH mapping population was developed from a cross

between 2 parents of T. aestivum, a winter wheat cultivar‘Arina’ and NK93604 (a Norwegian spring wheat breedingline) using the wheat × maize system (Laurie and Bennett1988). ‘Arina’ was released in 1981 and has covered morethan 40% of the wheat acreage of Switzerland since 1985. Ithas excellent spike resistance against leaf and glume blotchcaused by Stagonospora nodorum, and a good level of resis-tance to Fusarium head blight (FHB) (Paillard et al. 2003).‘Arina’ was chosen as a donor parent owing to its good levelof resistance to FHB and to its greater ability to adapt to theScandinavian climatic conditions than other exotic hexaploidwheat cultivars. NK93604 is a well-adapted and highly pro-ductive breeding line in Norway. Genomic DNA of the par-ents and 93 DH lines was extracted from young leaves usingthe DNeasy Plant DNA extraction kit (Qiagen, Mississauga,Ont.).

AFLP assayAFLP analysis was carried out according to the methods

of Vos et al. (1995) with some modifications as described byGrønnerød et al. (2002). MseI and PstI were used asfrequent- and rare-cutter enzymes, respectively. Selectiveamplification was performed with primers having 3 selectivenucleotides, with the PstI primers labeled with 6-carboxy-fluorescein (6-FAM), hexachloro-6-carboxy-fluorescein(HEX), or tetrachloro-6-carboxy-fluorescein (TET), whereasthe MseI primers were unlabelled. Selective amplificationwas performed in 10 µL reaction volumes that contained0.25 µmol/L of each primer, 200 µmol/L each dNTP(Amersham Biosciences Europe GmbH, Freiburg, Ger-many), 1× PCR buffer (50 mmol/L KCl, 1.5 mmol/L MgCl2,10 mmol/L Tris–HCl (pH 9.0)), 0.5 U Taq DNA polymerase(Amersham Biosciences), and 2.5 µL template DNA. The se-lective amplification was carried out using a “touchdown”profile of 1 cycle at 94 °C for 30 s, 65 °C for 30 s, and72 °C for 1 min, followed by 9 cycles over which the an-nealing temperature was decreased by 1 °C/cycle, and a finalstep of 23 cycles of 94 °C for 30 s, 56 °C for 30 s, and72 °C for 1 min. All PCRs were carried out using an MJResearch Tetrad thermal cycler (MJ Research, Waltham,Mass.).

AFLP fragments were separated on an ABI PRISM™ 377DNA Sequencer (PE Applied Biosystems, Foster City,Calif.) using 4.5% polyacrylamide denaturing gels (acryl-amide:bisacrylamide, 19:1) as described in the user manual.GS-500 (TAMRA labeled) size standard from ABI wasloaded into each lane to facilitate the semiautomatic analysisof the gel and the sizing of the fragments. Semiautomated

© 2006 NRC Canada

546 Genome Vol. 49, 2006

© 2006 NRC Canada

Semagn et al. 547

AFLP fragment analysis was performed with GeneScan™3.1 and Genotyper™ 2.1 programs as described in the usermanual, with some modification as reported by Schwarz etal. (2000).

Ninety-six MseI and PstI AFLP primer combinations werescreened for polymorphism and 45 informative primer pairswere used to genotype the mapping population. The nomen-clature for an AFLP marker is derived from the enzymecombination, the primer combinations, and the molecularmass of the product. The standard list for AFLP primer no-menclature provided by KeyGene (http://wheat.pw.usda.gov/ggpages/keygeneAFLPs.html) was employed.

Microsatellite assayA total of 874 wheat microsatellite (SSR) primers origi-

nated from 5 different groups were screened for polymor-phism: 150 BARC primers from the Beltsville AgriculturalResearch Center (Song et al. 2002, 2005), 21 CFA and 50CFD primers from INRA Clermont-Ferrand (Guyomarc’h etal. 2002; Sourdille et al. 2003), 195 DuPw primers fromDuPont company (Eujayl et al. 2002; DuPont, unpublisheddata; Dreisigacker et al. 2005), 50 GDM and 232 GWMprimers from Institut fur Pflanzengenetik und Kulturpflanzen-forschung (Roder et al. 1998a; Pestsova et al. 2000), and176 WMC primers from the Wheat Microsatellite Consor-tium (Gupta et al. 2002). Three hundred polymorphic prim-ers synthesized by Invitrogen Life Technologies (Carlsbad,Calif.) were used for segregation data collection. Micro-satellite analysis was performed using the fluorescent frag-ment detection system on an ABI PRISM 377 DNASequencer. For this method, either the forward primer wasdirectly labeled with a fluorochrome or had a 19-bp M13 tailat the 5′ end. In the latter case, a third universal fluorescent-labeled M13 primer (5′-CACGACGTTGTAAAACGAC-3′)was added to the PCR (Schuelke 2000; Somers et al. 2004).The PCR for directly labeled forward primers contained25 ng DNA, 1× PCR buffer, 0.30 µmol/L reverse primer,0.30 µmol/L forward primer labeled with 6-FAM, TET, orHEX, 200 µmol/L each dNTP, and 0.5 U Taq DNA polymer-ase (Amersham Biosciences Europe GmbH). For M13-tailedforward primers, the directly labeled forward primer was re-placed with 0.03 µmol/L M13 tailed forward primer and0.27 µmol/L of a universal fluorescent labeled M13 primer.Amplifications were performed for each primer pair sepa-rately in microtiter plates using the following programs:2 min at 94 °C, followed by 40 cycles of 1 min 94 °C, 1 minannealing (between 50 and 62 °C, depending on the pub-lished optimal annealing temperature of the primer), and1 min at 72 °C, and a final extension of 10 min at 72 °C.

SSR amplification products were multiplexed by combin-ing 0.7 µL of 6-FAM-, 0.7 µL of TET-, and 1.0 µL of HEX-labeled PCR products with 0.3 µL of GS-500 (TAMRA) and0.8 µL of ABI loading buffer (blue dextran and EDTA). Themultiplex was denatured for 3 min at 94 °C and quicklychilled on ice, and 0.8 µL of the denatured sample wasloaded on to a 4.5% denaturing polyacrylamide gel using amembrane-loading comb. SSR fragments were analyzed inthe same way as described for the AFLPs.

DArT assayDArT markers were generated by Triticarte Pty. Ltd.

(Canberra, Australia; http://www.triticarte.com.au), which isa whole-genome profiling service laboratory, as describedby Wenzl et al. (2004) and Akbari et al. (2006). Briefly, agenomic representation of a mixture of 13 cultivars was pro-duced after PstI–TaqI digestion, spotted on microarray slidesand the individual genotypes screened for polymorphismbased on fluorescence signals. The system was optimized forpolymorphism and detection accuracy. DNA from the par-ents (‘Arina’ and NK93604) was first screened for polymor-phism (giving a polymorphism rate of about 30%) and thenthe individual DH lines were genotyped. Two hundredseventy-one loci were scored as present (1) or absent (0).The locus designations used by Triticarte Pty. Ltd. wereadopted in this paper. DArT markers consisted of the prefix“wPt”, followed by numbers corresponding to a particularclone in the genomic representation, where w stands forwheat, P for PstI (primary restriction enzyme used) and t forTaqI (secondary restriction enzyme).

Segregation analysis and map constructionFor each segregating marker, a χ2 goodness-of-fit analysis

was performed to test for deviation from the 1:1 expectedsegregation ratio at a 1% level of significance. Linkagegroups were established using a minimum LOD score of 4.0after preliminary analysis using LOD scores ranging from 2to 10. A linkage group was assigned to a chromosome whenit contained at least 2 SSR loci that had been assigned to aparticular chromosome in previously published genetic (Liuand Tsunewaki 1991; Gale et al. 1995; Roder et al. 1998a;Li et al. 1999; Messmer et al. 1999; Mingeot and Jacquemin1999; Gupta et al. 2002; Paillard et al. 2003; Somers et al.2004; Song et al. 2005) and physical (Roder et al. 1998b;Sourdille et al. 2004; Song et al. 2005) maps. Final mappingwas done by combining 2 or more linkage groups that be-long to the same chromosome. The order of the markers ofeach chromosome was determined using fixed order of se-lected SSR loci from previous publications using the follow-ing parameters: recombination threshold = 0.45; minimumLOD score for calculating map distance = 0.10; ripple value =1; jump threshold = 5; and the Haldane mapping function.When necessary, markers were removed and the order recal-culated until a stable order was reached. All χ2 analysis andlinkage mapping were performed using JoinMap, v. 3.0 (VanOoijen and Voorrips 2001). The final map was drawn usingthe MapChart program, v. 2.1 (Voorrips 2002).

Results

Map constructionTwo hundred ninety polymorphic loci were scored from

45 AFLP primer pairs, an average of 6.4 loci/primer pair.The 300 SSR primers revealed 343 segregating loci that var-ied from 1 to 3 loci. Among the 904 loci (271 DArTs, 290AFLPs, and 343 SSRs) used for preliminary mapping in the‘Arina’ × NK93604 (ArNK) population, 724 loci mapped to58 linkage groups, each with 3–40 loci. The number of link-age groups remained basically the same when only the 343SSR markers were used for linkage analysis. Final mappingwas performed by combining 2 or more linkage groups thatbelong to the same chromosome, and a total of 280 non-informative loci (31%) were excluded from mapping for the

following reasons: (i) they did not meet the threshold valuefor entry with respect to jumps in goodness of fit; (ii) theycontributed to negative distance in the final map; and (or)(iii) they changed the order of anchoring markers.

The final genetic map (Fig. 1) consisted of 624 markersassigned to the 21 chromosomes, giving a total map lengthof 2595.5 cM. The number of SSRs with physical map infor-mation varied from 2 in 5B to 11 in 2B and 7A (markers inboldface in Fig. 1). The total number of mapped loci perchromosome ranged from 8 on 7D to 55 on 7A, and theaverage was 29.7 loci/chromosome. The largest chromosomemap was for 7A (202.8 cM); the shortest was for 4D(69.3 cM), and the average chromosome length was123.6 cM. The density of markers on the maps ranged from2.5 cM/marker on 6B to 12.2 cM/marker on 7D, with an av-erage density of 4.2 cM/marker. Map distances between 2consecutive markers varied from 0 to 49 cM, and 589 of the603 intervals (97.6%) were less than 20 cM. There wereonly 11 intervals with gaps larger than 20 cM, and the larg-est gaps between markers were observed on chromosomes6A (42.4 cM) and 7D (49.0 cM).

There was variation in the number of markers, maplength, and marker density on the basis of the homoeologousgroup. Marker number and density were the highest inhomoeologous group 2 (117 loci and 3.5 cM/marker, respec-tively), whereas map length was the highest (423.8 cM) inboth groups 1 and 7. Homoeologous group 5 had the lowestnumber of markers (52 loci), lowest marker density(5.2 cM/marker), and shortest map length (269.4 cM).Homoeologous group 2 contained the highest number ofAFLPs and SSRs, whereas DArTs were highest in group 7.For the former, the number of AFLPs was more than twicethat of the number of DArTs.

Map length was relatively shorter for the D genome(813.8 cM) than for the A (929.0) and B (852.7 cM)genomes. The distribution of markers among the genomeswas also not uniform, with twice as many markers mappedto the A and B genomes than the D genome: 245, 257, and122 markers for the A, B, and D genomes, respectively. Thenumber of AFLP and DArT markers mapped to the D ge-nome was very low (27 AFLPs and 20 DArTs) comparedwith the A (71 loci for each) and B (67 AFLPs and 98DArTs) genomes.

Segregation distortionOn the whole, the ArNK population was not skewed, with

49.5% for the alleles coming from ‘Arina’ and 51.5% of thealleles from NK93604. However, χ2 segregation tests foreach locus showed significant (p < 0.01) segregation distor-tion for 127 markers (20.4%). Markers exhibiting segrega-tion distortion in favor of NK93604 alleles were morefrequent (56.7%) than those in favor of ‘Arina’ alleles(43.3%). About 60% of the entire length of chromosome 2Dshowed distorted segregation, with an excess of NK93604alleles. Seven other chromosomes showed minor regions ofdistorted segregation. In this study, a chromosomal regionwas regarded as being associated with skewed segregationif 4 or more closely linked markers showed significantsegregation distortion. Based on these criteria, 10 regions ofsignificant segregation distortion (shown as filled or cross-hatched boxes within chromosomes in Fig. 1) were found on

7 chromosomes: 6 regions on chromosomes 1A, 3A, 4D,5D, and 6B contained an excess of alleles from ‘Arina’, and3 other regions on 2A and 7A contained an excess of allelesfrom NK93604.

Marker evaluationAmong the 624 markers mapped to the 21 chromosomes,

the majority of markers were SSRs (270), with 165 AFLPsand 189 DArTs. Chromosomes 2A, 7A, and 7B have thehighest number of AFLPs, SSRs, and DArTs, respectively. Atotal of 40 AFLPs (24.2%), 26 DArTs (13.8%), and 61 SSRs(22.6%) showed significant (p < 0.01) segregation distortion.Clustering of intra- and inter-marker intervals below a mapdistance of 0.5 cM was observed in 110 of 603 intervals(18.2%) between adjacent markers. The total number of in-tervals between 2 consecutive loci of the same marker typewas 52 for AFLPs, 90 for DArTs, and 131 for SSRs,whereas the remaining 330 intervals were between differentmarker types. Map distances in 51 intervals between 2 con-secutive DArT markers (27.0%) was less than 0.5 cM, whichwas much higher compared with 3 for AFLP (1.8%) and 24for SSR (8.9%). Clustering of DArTs was more frequent onchromosome 3B, 3D, 4A, and 7A, and was more frequentfor SSRs on 2A and 4A.

Discussion

Linkage groups and marker orderPreliminary linkage mapping in the present study identi-

fied 58 linkage groups, many more than the 21 haploid chro-mosomes of hexaploid wheat. This large number of linkagegroups compared with haploid chromosome number maysuggest that several areas of the genome remain undetectedwith the present set of markers. The requirement for a largenumber of markers or mapping populations to reduce thelinkage group number to haploid chromosome numbers andincrease map accuracy has been emphasized in other map-ping studies (e.g., Jeuken et al. 2001; Sharma et al. 2002;Crane and Crane 2005). To test if the large number of link-age groups in the ArNK population was due to poor dataquality, errors in scoring or establishing linkage groups, and(or) insufficient coverage of the genome by the markers, theraw segregation data (1468 loci for 115 recombinant inbredlines) for the most extensively studied ITMI population(Wheat, Synthetic × ‘Opata’, BARC) was downloaded fromthe GrainGenes Web site (http://wheat.pw.usda.gov/GG2/index.shtml), and mapped in the same way as described bySong et al. (2005). In contrast to the 21 groups reported bythe authors from the same dataset, we found 56–63 linkagegroups depending on the LOD score used to establish link-ages. For the latter, the number of linkage groups per chro-mosome varied from 1 for 4B and 4D to 5 for 7A, 7B, and7D. The authors have also confirmed the presence of severallinkage groups for several chromosomes during preliminarymapping of the ITMI population (J.R. Shi and Q.J. Song,personal communication). Twenty-one linkage groups withthe same map order and length as that of Song et al. (2005)was obtained only when 2 or more linkage groups that be-long to the same chromosome were combined and remappedusing selected anchoring markers, based on previous publi-

© 2006 NRC Canada

548 Genome Vol. 49, 2006

© 2006 NRC Canada

Semagn et al. 549F

ig.

1.G

enet

icli

nkag

em

apof

624

loci

in93

doub

le-h

aplo

idli

nes

deri

ved

from

acr

oss

betw

een

Trit

icum

aest

ivum

Lem

.T

hell

‘Ari

na’

and

NK

9360

4.A

FL

Pm

arke

rsar

ena

med

asde

scri

bed

inM

ater

ials

and

met

hods

.T

hebl

ue,

gree

n,an

dbl

ack

colo

rsin

dica

teA

FL

P,

DA

rT,

and

SS

Rm

arke

rs,

resp

ecti

vely

.A

sing

leS

SR

prim

erw

ith

2m

appe

dlo

cion

the

sam

ech

rom

osom

eha

vea

suff

ixof

eith

er“.

1”or

“.2”

.S

SR

sth

atsh

owed

disc

repa

ncie

sin

term

sof

chro

mos

ome

orm

arke

ror

der

com

pare

dw

ith

publ

ishe

dm

aps

are

inpa

rent

hese

s.S

SR

sw

ith

avai

labl

eph

ysic

alm

appi

ngin

form

atio

nar

ein

bold

face

.S

SR

skn

own

bych

rom

osom

allo

cati

onw

itho

utsp

ecif

icm

arke

ror

der

are

init

alic

s.C

umul

a-ti

vem

apdi

stan

ces

are

show

nin

cent

imor

gans

(cM

)on

the

left

side

ofth

eli

nkag

egr

oups

and

was

calc

ulat

edus

ing

the

Hal

dane

map

ping

func

tion

.L

oci

mar

ked

wit

h*

and

**de

viat

esi

gnif

ican

tly

from

1:1

rati

oat

p<

0.01

and

p<

0.00

1,re

spec

tive

ly.

Chr

omos

ome

segm

ents

show

nin

shad

edbo

xes

indi

cate

regi

ons

ofsi

gnif

ican

t(p

<0.

01)

segr

egat

ion

dist

orti

on(c

ross

hatc

hed

box,

‘Ari

na’

alle

les

inex

cess

;so

lid

fill

edbo

x,N

K93

604

alle

les

inex

cess

).

© 2006 NRC Canada

550 Genome Vol. 49, 2006F

ig1

(con

tinu

ed).

© 2006 NRC Canada

Semagn et al. 551F

ig1

(con

clud

ed).

cations, as fixed order. We therefore followed the same ap-proach to construct the final map for the ArNK population.

SSR marker order from the present study was comparedwith published maps in hexaploid wheat. Although there arevarious published maps for all hexaploid wheat chromo-somes (e.g., Liu and Tsunewaki 1991; Gale et al. 1995;Roder et al. 1998a; Li et al. 1999; Messmer et al. 1999;Mingeot and Jacquemin 1999; Gupta et al. 2002; Paillard etal. 2003; Somers et al. 2004; Song et al. 2005), the authorsbelieve that the wheat consensus map (Somers et al. 2004),the ITMI map (Song et al. 2005), and the wheat compositemap by Rudi Appels (http://www.shigen.nig.ac.jp/wheat/komugi/maps/markerMap.jsp) are the most detailed maps forcomparative purposes. The presence of various contradictorylocus orders among these maps, as detected by the freelyavailable CMap program at the GrainGenes Web site, cre-ated difficulty in deciding the correct locus order. For theArNK population, different locus orders were checked andthe best orders were compared with one or more of thesemaps. SSR marker order correlated very well with the con-sensus map in 10 chromosomes (1D, 2B, 2D, 3A, 4D, 5A,5B, 5D, 6B, and 7D) and with the composite map in 2 chro-mosomes (2A and 3D). Marker order in 1B was the same asthe consensus map in all except the interval betweenXgwm11 and Xbarc187, which was in agreement with theITMI map. On 7A, the interval from Xgwm282 toXgwm525 was the same as the consensus map, but the otherwas the same as the ITMI map. SSR marker orders in theother chromosomes showed partial correlation to 2 or 3 ofthese maps at different regions: 3 chromosomes (3B, 4B,and 7B) to the consensus and ITMI maps, 2 chromosomes(4A and 6A) to the consensus and composite maps, and 2other chromosomes (1A and 6D) to 3 maps at different re-gions. A difference in marker order among genetic maps isnot unexpected, as genetic mapping only gives an indicationof the relative position of the markers to each other (Sourdilleet al. 2004). The chromosomal location of 38 SSRs (loci initalic in Fig. 1) was available at the GrainGenes Web site, buttheir map position and locus order in their respective chromo-somes was the first in the present study.

Genome coverageThe ArNK map covers 2595.5 cM, which is similar to the

2569 cM of the wheat consensus map (Somers et al. 2004)and 2655 cM of the recently published ITMI map (Song etal. 2005), but it represents only 84% of 3086 cM map in theArFo (Paillard et al. 2003). The ArNK map shows 7 chro-mosomes (1A, 1D, 4A, 4B, 4D, 6B, and 6D) as being lon-ger, and shows the remaining 14 chromosomes as beingshorter. In fact, the ArFo map length for 11 chromosomes(1B, 2A, 2B, 2D, 3B, 3D, 5A, 5D, 6A, 7A, and 7D) wasmuch longer than that of the high-density wheat consensusand ITMI maps. The overall average marker density in theArNK map (4.2 cM/marker) was much better than that of theArFo maps (8 cM/marker), but lower than that of the ITMI(1.9 cM/marker) and the wheat consensus (2.2 cM) maps.Despite the larger number of markers in the present map, thetotal genome coverage and the length of several individualchromosomes are shorter than the ArFo map. Differences inrecombination frequencies owing to genetic background ef-

fects from the other parents and (or) a difference in mappingalgorithm may be possible explanations for such differences.

Although we generally had good coverage, the short armsof 5 chromosomes (3A, 4A, 4B, 4D, and 6A) and the longarm of 5D were under represented in the map. In all except4BS, this was mainly due to lack of a sufficient number ofpolymorphic microsatellite markers in the correspondingregions (Somers et al. 2004; Song et al. 2005). The totallength and the number of markers mapped to the D genomewere much lower than the A and B genomes. The low levelof polymorphism in the D genome is well known (e.g., Röderet al. 1998a; Chalmers et al. 2001) and is in agreement withthe hypothesis of a mono- or oligophyletic origin of hexaploidwheat (Lagudah et al. 1991; Salamini et al. 2002).

Segregation distortionThe observed segregation distortion (20.4%) in the ArNK

population is in agreement with Cadalen et al. (1997), whofound segregation distortion of 20% at a 1% level of signifi-cance. Markers showing distorted segregation were clusteredtogether on specific regions on 8 chromosomes (1A, 2A, 2D,3A, 4D, 5D, 6B, and 7A). Clustering of distorted markers onvarious chromosomes have been reported in various studies(Cadalen et al. 1997; Faris et al. 1998; Messmer et al. 1999;Kammholz et al. 2001; Paillard et al. 2003), and it may becaused by different factors. First, the ArNK population wasderived from a wide cross between winter and spring wheat,and segregation distortion is a common phenomenon in widecrosses and has been reported on various wheat chromo-somes (Cadalen et al. 1997; Messmer et al. 1999; Kammholzet al. 2001; Paillard et al. 2003). Second, the ArNK popula-tion was also developed by means of wheat × maize crosses,and non-Mendelian segregation within wheat × maize de-rived DH populations may be due to (i) heterogeneity withinthe parents, (ii) selection associated with the DH productionprocess, and (iii) outcrossing and admixture of seed duringincrease for trials (Kammholz et al. 2001). Third, game-tocidal and pollen-killer genes have been reported in wheatas being distributed in different homoeologous groups (Gillet al. 1996a; Faris et al. 1998a; McIntosh et al. 1998) andmay be responsible for the observed distorted segregationin the ArNK population. Kammholz et al. (2001) had alsospeculated that clustering of distorted loci appeared to beassociated with introgressed alien chromatin involvingtranslocations into one of the parents. However, the parentsused in this study are not known to have any alien trans-locations.

Distribution of markers among and within linkagemaps

In some chromosomes (e.g., 2A and 2D), AFLP markersappeared as clustered but the interval between most adjacentmarkers was >1.5 cM. In contrast to frequent clustering ofAFLP markers reported by several researchers at thecentromere and surrounding heterochromatin regions of thegenome based on an EcoRI–MseI combination, extensiveclustering was not apparent in the ArNK genetic map. Themethylation-sensitive PstI–MseI combination was used inthe present study, which revealed a significantly lower levelof polymorphism in wheat than the methylation-insensitiveEcoRI–MseI combination (Barrett and Kidwell 1998). Co-

© 2006 NRC Canada

552 Genome Vol. 49, 2006

migration and scoring error may, therefore, be very muchlower in the PstI–MseI than in the EcoRI–MseI combination.AFLP fragments were analyzed using an ABI PRISM 377DNA Sequencer in combination with the computer softwareGenScan and Genotyper (PE Applied Biosystems), and theapplication of an internal-lane size standard ensured accuratesizing of fragments by minimizing the lane-to-lane or gel-togel variation present in other studies that used silver stainingor radioactivity, as reported by Huang et al. (2000) andSchwarz et al. (2000).

Clustering of SSR markers was observed in 3 chromo-somes (1B, 2A, and 4B). Roder et al. (1998a) reported clus-tering of microsatellites in several centromeric regions on 6chromosomes in hexaploid wheat. Gene-rich regions in theeuchromatin tend to be low copy and hypomethylated(Naveh-Many and Cedar 1981). Cardle et al. (2000) andMorgante et al. (2002) indicated that microsatellites are sig-nificantly associated with the low-copy fraction of plantgenomes based on the estimation of their density in Arabi-dopsis, rice, soybean, maize, and wheat. Clustering may,therefore, occur when genetic maps are constructed withhypomethylated and low-copy SSRs.

The frequency of clustering in DArT markers at intervalsless than 0.5 cM was 3- and 15-fold higher than SSRs andAFLPs, respectively (Fig. 2). High-density physical maps inwheat reveal that more than 85% of wheat genes are presentin gene-rich regions, physically spanning only 5%–10% ofthe chromosomal regions (Gill et al. 1996a, 1996b; Faris etal. 2000; Sandhu et al. 2001). A high correlation betweengene-rich regions and recombination is common in wheat(Gill et al. 1996a, 1996b; Sandhu et al. 2001; Weng andLazar 2002) and many other organisms. In homoeologousgroup 6 chromosomes, for example, gene-rich regions pres-

ent near telomeric ends show 10–20 fold more recombina-tion than proximal regions on the chromosome (Shah andHassan 2005). The high proportion of clustering of DArTmarkers may therefore be indicative of gene-rich regions.

Clustering in DArTs may also be due to the presence ofredundant clones in the genomic representation, which wassupported in the following 2 ways: (i) all intervals <0.5 cMbetween 2 consecutive DArT markers showed high correla-tion coefficients (r > 0.98) and (ii) strong multicollinearitieswere observed between DArT markers with such high corre-lation coefficients in QTL analysis for FHB resistance (datanot shown). Genomic reduction is the critical step in produc-ing genomic representations with a sufficient number ofunique polymorphic clones in DArT libraries. This step maybe more complex in species with high ploidy level and ge-nome size, such as hexaploid wheat. The data from the pres-ent study may, therefore, suggest that the present genomicrepresentations for hexaploid wheat at Triticarte Pty. Ltd. re-quires additional sieving steps to exclude highly correlatedclones and incorporate more unique clones with even distri-bution along chromosomes.

The map presented here provides insight regarding thedistribution of DArT markers in comparison with the well-established microsatellite and AFLP markers for linkagemapping. It also provided readily detectable markers foridentifying novel quantitative trait loci for resistance to FHBand powdery mildew and for anther extrusion, leaf segmentincubation, and latency.

Acknowledgements

We thank Dr. M. Roder (IPK Gatersleben, Germany) forproviding some unpublished primer sequences. This work

© 2006 NRC Canada

Semagn et al. 553

Fig. 2. Observed distribution of Haldane map distance between 2 consecutive loci over all chromosomes. AFLP, DArT, and SSR areintramarker intervals. Bars represent standard error of the mean.

was supported by the Research Council of Norway (projectNo. 151281/110) and was a collaboration with CIMMYT.

References

Akbari, M., Wenzl, P., Caig, V., Carlig, J., Xia, L., Yang, S., et al.2006. Diversity arrays technology (DArT) for high-throughputprofiling of the hexaploid wheat genome. Theor. Appl. Genet. Inpress.

Akkaya, M.S., Bhagwat, A.A., and Cregan, P.B. 1992. Lengthpolymorphism of simple sequence repeat DNA in soybean. Ge-netics, 132: 1131–1139.

Barrett, B.A., and Kidwell, K.K. 1998. AFLP-based genetic diver-sity assessment among wheat cultivars from the Pacific North-west. Crop Sci. 38: 1261–1271.

Bennett, M.D., and Smith, J.B. 1976. Nuclear DNA amounts in an-giosperms. Philos. Trans. R. Soc. Lond. Ser. B. Biol. Sci. 274:227–274.

Botstein, D., White, R.L., Skolnick, M., and Davis, R.W. 1980.Construction of a genetic linkage map in man using restrictionfragment length polymorphisms. Am. J. Hum. Genet. 32: 314–331.

Cadalen, T., Boeuf, C., Bernard, S., and Bernard, M. 1997. Anintervarietal molecular marker map in Triticum aestivum L. Em.Thell., and comparison with a map from a wide cross. Theor.Appl. Genet. 94: 367–377.

Cardle, L., Ramsay, L., Milbourne, D., Macaulay, M., Marshal, D.,and Waugh, R. 2000. Computational and experimental charac-terization of physically clustered simple sequence repeats inplants. Genetics, 156: 847–854.

Chalmers, K.J., Cambell, A.W., Kretschmer, J., Karakousis, A.,Henschke, P.H., Pierens, S., et al. 2001. Construction of threelinkage maps in bread wheat, Triticum aestivum. Aust. J. Agric.Res. 52: 1089–1119.

Crane, C.F., and Crane, Y.M. 2005. A nearest-neighboring-end al-gorithm for genetic mapping. Bioinformatics, 21: 1579–1591.

Dreisigacker, S., Zhang, P., Warburton, M.L., Skovmand, B.,Hoisington, D., and Melchinger, A.E. 2005 Genetic diversityamong and within CIMMYT landrace accessions investigatedwith SSRs and implications for plant genetic resources manage-ment. Crop Sci. 45: 653–661.

Eujayl, L., Sorrells, M.E., Baum, M., Wolters, P., and Powell, W.2002. Isolation of EST-derived microsatellite markers for geno-typing the A and B genomes of wheat. Theor. Appl. Genet. 104:399–407.

Faris, J.D., Laddomada, B., and Gill, B.S. 1998. Molecular map-ping of segregation distortion loci in Aegilops tauschii. Genet-ics, 149: 319–327.

Faris, J.D., Haen, K.M., and Gill, B.S. 2000. Saturation mappingof a gene-rich recombination hot spot region in wheat. Genetics,154: 823–835.

Gale, M.D., Atkinson, M.D., Chinoy, C.N., Harcourt, R.L., Jia, J.,Li, Q.Y., and Devos, K.M. 1995. Genetic maps of hexaploidwheat. Proceedings of the 8th International Wheat GeneticsSymposium. 20–25 July 1993, Beijing, China. Edited by Z.S.Li and Z.Y. Xin. China Agricultural Scientech Press, Vol. 1.pp. 29–40.

Gill, K.S., Gill, B.S., Endo, T.R., and Boyko, E.V. 1996a. Identifi-cation of high-density mapping of gene-rich regions in chromo-some group 5 of wheat. Genetics, 143: 1001–1012.

Gill, K.S., Gill, B.S., Endo, T.R., and Taylor. T. 1996b. Identifica-tion and high-density mapping of gene-rich regions in chromo-some group 1 of wheat. Genetics, 144: 1883–1891.

Grønnerød, S., Marøy, A.G., MacKey, J., Tekauz, A., Penner, G.A.,and Bjørnstad, A. 2002. Genetic analysis of resistance to barleyscald (Rhynchosporium secalis) in the Ethiopian line ‘Abyssin-ian’ (CI668). Euphytica, 126: 235–250.

Gupta, P.K., Varshney, R.K., Sharma, P.C., and Ramesh, B. 1999.Molecular markers and their application in wheat breeding: a re-view. Plant Breed. 118: 369–390.

Gupta, P.K., Balyan, H.S., Edwards, K.J., Isaac, P., Korzun, V.,Röder, M., et al. 2002. Genetic mapping of 66 new micro-satellite (SSR) loci in bread wheat. Theor. Appl. Genet. 105:413–422.

Guyomarc’h, H., Sourdille, P., Charmet, G., Edwards, K.J., andBernard, M. 2002. Characterization of polymorphic micro-satellite markers from Aegilops tauschii and transferability tothe D-genome of bread wheat. Theor. Appl. Genet. 104: 1164–1172.

Huang, X., Zeller, F.J., Hsam, S.L.K., Wenzel, G., and Mohler, V.2000. Chromosomal location of AFLP markers in commonwheat utilizing nulli-tetrasomic stocks. Genome, 43: 298–305.

Jaccoud, D., Peng, K., Feinstein, D., and Kilian, A. 2001. Diversityarrays: a solid state technology for sequence information inde-pendent genotyping. Nucleic Acids Res. [online], 29: e25. Avail-able from http://nar.oxfordjournals.org/cgi/reprint/29/4/e25.

Jeuken, M., van Wijk, R., Peleman, J., and Lindhout, P. 2001. Anintegrated interspecific AFLP map of lettuce (Lactuca) based ontwo L. sativa × L. saligna F2 populations. Theor. Appl. Genet.103: 638–647.

Kammholz, S.J., Campbell, A.W., Sutherland, M.W., Hollamby,G.J., Martin, P.J., Eastwood, R.F., et al. 2001. Establishment andcharacterization of wheat genetic mapping populations. Aust. J.Agric. Res. 52: 1079–1088.

Lagudah, E.S., Appels, R., Brown, A.H.D., and McNeil, D. 1991.The molecular-genetic analysis of Triticum tauschii, the D-genome donor to hexaploid wheat. Genome, 34: 375–386.

Laurie, D.A., and Bennett, M.D. 1988. The production of haploidwheat plants from wheat × maize crosses. Theor. Appl. Genet.76: 393–397.

Li, W.L., Faris, J.D., Chittoor, J.M., Leach, J.E., Hulbert, S.H., Liu,D.J., et al. 1999. Genomic mapping of defense response genesin wheat. Theor. Appl. Genet. 98: 226–233.

Liu, Y.G., and Tsunewaki, K. 1991. Restriction fragment lengthpolymorphism (RFLP) analysis in wheat. II. Linkage maps ofthe RFLP sites in common wheat. Jpn. J. Genet. 66: 617–633.

Marino, C.L., Nelson, J.C., Lu, Y.H., Sorrells, M.E., Leroy, P.,Tuleen, N.A., et al. 1996. Molecular genetic maps of the group 6chromosomes of hexaploid wheat (Triticum aestivum L. em.Thell). Genome, 39: 359–366.

McIntosh, R.A., Hart, G.E., Devos, K.M., Gale, M.D., and Rogers,W.J. 1998. Catalogue of gene symbols for wheat. In Proceedingsof the 9th International Wheat Genetics Symposium, 2–7 Au-gust 1998, Saskatoon, Saskatchewan. Slinkard (ed.). UniversityExtension Press. Vol. 5. pp. 13–72.

Messmer, M.M., Keller, M., Zanetti, S., and Keller, B. 1999. Ge-netic linkage map of a wheat × spelt cross. Theor. Appl. Genet.98: 1163–1170.

Mingeot, D., and Jacquemin, J.M. 1999. Mapping of RFLP probescharacterized for their polymorphism on wheat. Theor. Appl.Genet. 98: 1132–1137.

Morgante, M., Hanafey, M., and Powell, W. 2002. Microsatellitesare preferentially associated with nonrepetitive DNA in plantgenomes. Nat. Genet. 30: 194–200.

Naveh-Many, T., and Cedar, H. 1981. Active gene sequences areundermethylated. Proc. Natl. Acad. Sci. U.S.A. 78: 4226–4250.

© 2006 NRC Canada

554 Genome Vol. 49, 2006

Nelson, J.C., van Deynze, A.E., Autrique, E., Sorrells, M.E., Lu,Y.H., Merlino, M., et al. 1995a. Molecular mapping of wheat.Homoeologous group 2. Genome, 38: 516–524.

Nelson, J.C., van Deynze, A.E., Autrique, E., Sorrells, M.E., Lu,Y.H., Negre, S., et al. 1995b. Molecular mapping of wheat.Homoeologous group 3. Genome, 38: 525–533.

Nelson, J.C., Sorrells, M.E., van Deynze, A.E., Lu, Y.H., Atinkson,M., Bernard, M., et al. 1995c. Molecular mapping of wheat. Ma-jor genes and rearrangements in homoeologous groups 4, 5, and7. Genetics, 141: 721–731.

Paillard, S., Schnurbusch, T., Winzeler, M., Messmer, M.,Sourdille, P., Abderhalden, O., et al. 2003. An integrative ge-netic linkage map of winter wheat (Triticum aestivum L.).Theor. Appl. Genet. 107: 1235–1242.

Parker, G.D., Chalmers, K.J., Rathjen, A.J., and Langridge, P.1998. Mapping loci associated with flour colour in wheat(Triticum aestivum L.). Theor. Appl. Genet. 96: 238–245.

Peleman, J.D., and van der Voort, J.R. 2003. Breeding by design.Trends Plant Sci. 8: 330–334.

Pestsova, E., Ganal, M.W., and Roder, M.S. 2000. Isolation andmapping of microsatellite markers specific for the D genome ofbread wheat. Genome, 43: 689–697.

Roder, M.S., Korzun, V., Wendehake, K., Plaschke, J., Tixier,M.H., Leroy, P., and Ganal, M.W. 1998a. A microsatellite mapof wheat. Genetics, 149: 2007–2023.

Roder, M.S., Korzun, V., Gill, B.S., and Ganal, M. 1998b. Thephysical mapping of microsatellite markers in wheat. Genome,41: 278–283.

Salamini, F., Ôzkan, H., Brandolini, A., Schäfer-Pregl, R., andMartin, W. 2002 Genetics and geography of wild cereal domes-tication in the Near East. Nat. Rev. Genet. 3: 429–441.

Sandhu, D., Champoux, J.A., Bondareva, S.N., and Gill, K.S.2001. Identification and physical localization of useful genesand markers to major gee-rich region on wheat group 1S chro-mosomes. Genetics, 157: 1735–1747.

Schuelke, M. 2000. An economic method for the fluorescent label-ing of PCR fragments. Nat. Biotechnol. 18: 233–234.

Schwarz, G., Herz, M., Huang, X.Q., Michalek, W., Jahoor, J.,Wenzel, G., and Mohler, V. 2000. Application of fluorescent-based semi-automated AFLP analysis in barley and wheat.Theor. Appl. Genet. 100: 545–551.

Shah, M.M., and Hassan, A. 2005. Distribution of genes recombi-nation on wheat homoeologous group 6 chromosomes: a synthe-sis of available information. Mol. Breed. 15: 45–53.

Sharma, R., Aggarwal, R.A.K., Kumar, R., Mohapatra, T., andSharma, R.P. 2002. Construction of an RAPD linkage map andlocalization of QTLs for oleic acid level using recombinantinbreds in mustard (Brassica juncea). Genome, 45: 467–472.

Somers, D.J., Isaac, P., and Edwards, K. 2004. A high-densitymicrosatellite consensus map for bread wheat (Triticumaestivum L.). Theor. Appl. Genet. 109: 1105–1114.

Song, Q.J., Fickus, E.W., and Cregan, P.B. 2002. Characterizationof trinucleotide SSR motifs in wheat. Theor. Appl. Genet. 104:286–293.

Song, Q.J., Shi, J.R., Singh, S., Fickus, E.W., Costa, J.M., Lewis,J., et al. 2005. Development and mapping of microsatellite(SSR) markers in wheat. Theor. Appl. Genet. 110: 550–560.

Sourdille, P., Cadalen, T., Guyomarc’h, H., Snape, J.W., Perretant,M.R., Charmet, G., et al. 2003. An update of the Courtot × Chi-nese Spring intervarietal molecular marker linkage map for theQTL detection of agronomic traits in wheat. Theor. Appl. Genet.106: 530–538.

Sourdille, P., Singh, S., Cadalen, T., Brown-Guedira, G.L., Gay, G.,Qi, L., et al. 2004. Microsatellite-based deletion bin system forthe establishment of genetic-physical map relationships in wheat(Triticum aestivum L.). Funct. Integr. Genomics, 4: 12–25.

Talbert, L.E., Blake, N.K., Chee, P.W., Blake, T.K., and Magyar,G.M. 1994. Evaluation of sequence-tagged-site-facilitated PCRproducts as molecular markers in wheat. Theor. Appl. Genet. 87:789–794.

Van Deynze, A.E., Dubcovsky, J., Gill, K.S., Nelson, J.C., Sorrells,M.E., Dvorak, J., et al. 1995. Molecular-genetic maps for group1 chromosomes of Triticeae species and their relation to chro-mosomes in rice and oat. Genome, 38: 45–59.

Van Ooijen, J.W., and Voorrips, R.E. 2001. JoinMap 3.0 softwarefor the calculation of genetic linkage maps [computer program].Plant Research International, Wageningen, the Netherlands.

Voorrips, R.E. 2002. MapChart: software for the graphical presen-tation of linkage maps and QTLs. J. Hered. 93: 77–78.

Vos, P., Hogers, R., Reijans, M., van de Lee, T., Hornes, M.,Friters, A., et al. 1995. AFLP: a new technique for DNA finger-printing. Nucleic Acids Res. 23: 4407–4414.

Weng, Y., and Lazar, M.D. 2002. Comparison of homoeologousgroup-6 short arm physical maps of wheat and barley reveals asimilar distribution of recombinogenic and gene-rich regions.Theor. Appl. Genet. 104: 1078–1085.

Wenzl, P., Carling, J., Kudrna, D., Jaccoud, D., Huttner, E.,Kleinhofs, A., and Kilian, A. 2004. Diversity arrays technology(DArT) for whole-genome profiling of barley. Proc. Natl. Acad.Sci. U.S.A. 101: 9915–9920.

Williams, J.G.K., Kubelik, A.R.K., Livak, J.L., Rafalski, J.A., andTingey, S.V. 1990. DNA polymorphisms amplified by randomprimers are useful as genetic markers. Nucleic Acids Res. 18:6531–6535.

Wittenberg, A.H.J., van der Lee, T., Cayla, C., Kilian, A., Visser,R.G.F., and Schouten, H.J. 2005. Validation of the high-throughput marker technology DArT using the model plantArabidopsis thaliana. Mol. Genet. Genomics, 274: 30–39.

Xia, L., Peng, K., Yang, S., Wenzl, P., de Vicente, M.C., Fregene,M., and Kilian, A. 2005. DArT for high-throughput genotypingof cassava (Manihot esculenta) and its wild relatives. Theor.Appl. Genet. 110: 1092–1098.

© 2006 NRC Canada

Semagn et al. 555