Development of a Near-Isogenic Line Population of
Copyrightó2007by the Genetics Society of America
DOI:10.1534/genetics.106.066423
Development of a Near-Isogenic Line Population of Arabidopsis thaliana and Comparison of Mapping Power With a Recombinant
Inbred Line Population
Joost J.B.Keurentjes,*,?Leo′nie Bentsink,*,1Carlos Alonso-Blanco,?Corrie J.Hanhart,* Hetty Blankestijn-De Vries,*Sigi Effgen,§Dick Vreugdenhil?and Maarten Koornneef*,§,2 *Laboratory of Genetics and?Laboratory of Plant Physiology,Wageningen University,NL-6703BD,Wageningen,The Netherlands,?Centro Nacional de Biotecnolog?′a(Consejo Superior de Investigaciones Cient?′?cas),28049Madrid,Spain and
§Max Planck Institute for Plant Breeding Research,50829Cologne,Germany
Manuscript received October5,2006
Accepted for publication November25,2006
ABSTRACT
In Arabidopsis recombinant inbred line(RIL)populations are widely used for quantitative trait locus
(QTL)analyses.However,mapping analyses with this type of population can be limited because of the
masking effects of major QTL and epistatic interactions of multiple QTL.An alternative type of immortal
experimental population commonly used in plant species are sets of introgression lines.Here we
introduce the development of a genomewide coverage near-isogenic line(NIL)population of Arabidopsis
thaliana,by introgressing genomic regions from the Cape Verde Islands(Cvi)accession into the
Landsberg erecta(L er)genetic background.We have empirically compared the QTL mapping power of
this new population with an already existing RIL population derived from the same parents.For that,we
analyzed and mapped QTL affecting six developmental traits with different heritability.Overall,in the
NIL population smaller-effect QTL than in the RIL population could be detected although the
localization resolution was lower.Furthermore,we estimated the effect of population size and of the
number of replicates on the detection power of QTL affecting the developmental traits.In general,
population size is more important than the number of replicates to increase the mapping power of RILs,
whereas for NILs several replicates are absolutely required.These analyses are expected to facilitate
experimental design for QTL mapping using these two common types of segregating populations.
Q UANTITATIVE traits are characterized by contin-uous variation.The establishment of the genetic basis of quantitative traits is commonly referred to as quantitative trait locus(QTL)mapping and has been hampered due to their multigenic inheritance and the often strong interaction with the environment.The principle of QTL mapping in segregating populations is based on the genotyping of progeny derived from a cross of distinct genotypes for the trait under study. Phenotypic values for the quantitative trait are then compared with the molecular marker genotypes of the progeny to search for particular genomic regions showing statistically signi?cant associations with the trait variation,which are then called QTL(B roman 2001;S late2005).Over the past few decades,the?eld has bene?ted enormously from the progress made in molecular marker technology.The ease by which such markers can be developed has enabled the generation of dense genetic maps and the performance of QTL mapping studies of the most complex traits(B orevitz and N ordborg2003).
QTL analyses make use of the natural variation pres-ent within species(A lonso-B lanco and K oornneef 2000;M aloof2003)and have been successfully applied to various types of segregating populations.In plants, the use of‘‘immortal’’mapping populations consisting of homozygous individuals is preferred because it allows performance of replications and multiple analyses of the same population.Homozygous populations can be obtained by repeated sel?ng,like for recombinant inbred lines(RILs),but also by induced chromosomal doubling of haploids,such as for doubled haploids (DHs)(H an et al.1997;R ae et al.1999;V on K orff et al. 2004).Depending on the species one can in principle also obtain immortality by vegetative propagation,al-though this is often more laborious.RILs are advanta-geous over DHs because of their higher recombination frequency in the population,resulting from multiple meiotic events that occurred during repeated sel?ng (J ansen2003).
Another type of immortal population consists of introgression lines(ILs)(E shed and Z amir1995), which are obtained through repeated backcrossing and extensive genotyping.These are also referred to
1Present address:Department of Molecular Plant Physiology,Utrecht University,NL-3584CH,Utrecht,The Netherlands.
2Corresponding author:Laboratory of Genetics,Wageningen University, Arboretumlaan4,NL-6703BD,Wageningen,The Netherlands.
E-mail:maarten.koornneef@wur.nl
Genetics175:891–905(February2007)
as near-isogenic lines(NILs)(M onforte and T anksley 2000)or backcross inbred lines(BILs)(J euken and L indhout2004;B lanco et al.2006).Such populations consist of lines containing a single fragment or a small number of genomic introgression fragments from a donor parent into an otherwise homogeneous genetic background.Although no essential differences exist be-tween these populations,we use the term near-isogenic lines for the materials described here.A special case of ILs are chromosomal substitution strains(CSSs) (N adeau et al.2000;K oumproglou et al.2002),where the introgressions span complete chromosomes.
All immortal populations except those that can be propagated only vegetatively share the advantage that they can easily be maintained through seeds,which allows the analysis of different environmental in?uences and the study of multiple,even invasive or destructive, traits.Statistical power of such analyses is increased because replicate measurements of genetically iden-tical individuals can be done.
In plants,RILs and NILs are the most common types of experimental populations used for the analysis of quantitative traits.In both cases the accuracy of QTL lo-calization,referred to as mapping resolution,depends on population size.For RILs,recombination frequency within existing lines is?xed and can therefore be increased within the population only by adding more lines(i.e.,more independent recombination events).Al-ternatively,recombination frequency can be increased by intercrossing lines before?xation by inbreeding as homozygous lines(Z ou et al.2005).In NIL populations resolution can be improved by minimizing the intro-gression size of each NIL.Consequently,to maintain genomewide coverage a larger number of lines are needed.Despite the similarities between these two types of mapping populations,large differences exist in the genetic makeup of the respective individuals and the resulting mapping approach.In general,recombina-tion frequency in RIL populations is higher than that in equally sized NIL populations,which allows the analysis of less individuals.Each RIL contains several introgres-sion fragments and,on average,each genomic region is represented by an equal number of both parental genotypes in the population.Therefore,replication of individual lines is often not necessary because the effect of each genomic region on phenotypic traits is tested by comparing the two genotypic RIL classes(each com-prising approximately half the number of lines in the population).In addition,the multiple introgressions per RIL allow detection of genetic interactions between loci(epistasis).However,epistasis together with un-equal recombination frequencies throughout the ge-nome and segregation distortions caused by lethality or reduced?tness of particular genotypes may bias the power to detect QTL.Furthermore,the wide variation of morphological and developmental traits present in most RIL populations may hamper the analysis of traits requiring the same growth and developmental stage of the individual lines.When many traits segregate simul-taneously,this often affects the expression of other traits due to genetic interactions.Moreover,large-effect QTL may mask the detection of QTL with a small additive effect.
In contrast to RILs,NILs contain only a single intro-gression per line,which increases the power to detect small-effect QTL.However,the presence of a single in-trogression segment does not allow testing for genetic interactions and thereby the detection of QTL ex-pressed in speci?c genetic backgrounds(epistasis).In addition,because most of the genetic background is identical for all lines,NILs show more limited develop-mental and growth variation,increasing the homoge-neity of growth stage within experiments.Nevertheless, lethality and sterility might sometimes hinder the obtaining of speci?c single introgression lines.
The choice of one mapping population over another depends on the plant species and the speci?c parents of interest.In cases where different cultivars or wild ac-cessions are studied preference is often given to RILs. However,when different species or when wild and cultivated germ plasm are combined(E shed and Z amir 1995;J euken and L indhout2004;V on K orff et al. 2004;B lair et al.2006;Y oon et al.2006)NILs are preferred.For instance,in tomato the high sterility in the offspring of crosses between cultivated and wild species made the use of NIL populations(E shed and Z amir1995)preferable because genomewide coverage cannot be obtained with RIL populations due to sterility, etc.Furthermore,the analysis of agronomical impor-tant traits(such as fruit characters)cannot be per-formed when many genes conferring reduced fertility segregate.In Arabidopsis,the easiness to generate fer-tile RIL populations with complete genome coverage, due to its fast generation time,has led to their extensive use in mapping quantitative traits.
NILs have been developed in various studies using Arabidopsis to con?rm and?ne map QTL previously identi?ed in RILs(A lonso-B lanco et al.1998b,2003; S warup et al.1999;B entsink et al.2003;E dwards et al. 2005;J uenger et al.2005a;T eng et al.2005)for which also heterogeneous inbred families(HIFs)(T uinstra et al.1997)have been used(L oudet et al.2005; R eymond et al.2006).A set of chromosomal substitu-tions of the Landsberg erecta(L er)accession into Colum-bia(Col)has been developed to serve as starting material for making smaller introgressions(K oumproglou et al. 2002).In mice CSSs are widely used for mapping purposes and have proven to be a valuable complement to other population types(S tylianou et al.2006). However,no genomewide set of NILs that allows map-ping to subparts of the chromosome has been described in Arabidopsis and,to our knowledge,no empirical comparative study has been performed between the two population types within a single species.
892J.J.B.Keurentjes et al.
In this study we aim to compare a RIL population with a NIL population in terms of QTL detection power and localization resolution.For that,we generated a new genomewide population of NILs using the same L er and Cape Verde Islands(Cvi)parental accessions as used earlier to generate a RIL population(A lonso-B lanco et al.1998a).The two experimental populations were grown simultaneously in the same experimental setup,including multiple replicates.QTL mapping analyses were performed on six different traits and the results of these analyses were compared in both populations.
MATERIALS AND METHODS
Mapping populations:Two types of mapping populations were used to analyze six developmental traits.The?rst population consists of a set of161RILs derived from a cross between the accessions Cvi and L er.The F10generation has been extensively genotyped(A lonso-B lanco et al.1998a)and is available from the Arabidopsis Biological Resource Center. All lines were advanced to the F13generation and residual heterozygous regions,estimated at0.71%in the F10genera-tion,were genotyped again with molecular PCR markers to con?rm that they were practically100%homozygous.
The second population consists of a set of92NILs.NILs were generated by selecting appropriate L er/Cvi RILs and repeated backcrossing with L er as recurrent female parent.
A number of these lines have been described previously
(A lonso-B lanco et al.1998b,2003;S warup et al.1999;
B entsink et al.2003;E dwards et al.2005;J uenger et al. 2005a;T eng et al.2005).The progeny of backcrosses was genotyped with PCR markers and lines containing a homozy-gous Cvi introgression into an otherwise L er background were selected.The set of selected lines was then extensively genotyped by AFLP analysis using the same restriction en-zymes and primer combinations as those used for the geno-typing of the RILs(A lonso-B lanco et al.1998a).The NILs will be made available through the Arabidopsis stock centers.
In both populations each line is almost completely homo-zygous and therefore individuals of the same line are genet-ically identical,which allows the pooling of replicated individuals and repeated measurements to obtain a more precise estimate of phenotypic values.For the RIL and NIL population16and24genetically identical plants were grown per line,respectively.Additionally,96replicates were grown for each parental accession L er and Cvi.All plants were grown in a single experiment with four completely random-ized blocks containing4,6,and24replicates per RIL,NIL, and parent,respectively.
Plant growing conditions:Seeds were sown in petri dishes on water-soaked?lter paper and incubated for5days in a cold room at4°in the dark to promote uniform germination. Subsequently,petri dishes were transferred to a climate chamber(24°,16hr light per day)for2days before planting. Germinated seedlings were transferred to clay pots,placed in peat,containing a sandy soil mixture.A single plant per pot was grown under long-day light conditions in an air-conditioned greenhouse from July until October.Plants were fertilized every2weeks using a liquid fertilizer. Quantitative traits:A total of six developmental traits,which were known to vary within the populations for the number of QTL and heritability,were measured on all individuals.We quanti?ed?owering time(FT);main in?orescence length at ?rst silique(SL);total length of the main in?orescence(TL);basal branch number(BB),which is the number of side shoots growing out from the rosette;main in?orescence branch number(IB),which is the number of elongated axillary (secondary)in?orescences along the main in?orescence; and total number of side shoots(TB)(basal plus main in?orescence).Flowering time was recorded as the number of days from the date of planting until the opening of the?rst ?ower.All other traits were measured at maturity. Quantitative genetic analyses:For both populations and for each trait,total phenotypic variance was partitioned into sources attributable to genotype(V G;i.e.,the line effect)and error(V E),using a random-effects analysis of variance (ANOVA,SPSS version11.0)according to the model y?m1G1E.Variance components were used to estimate broad sense heritability according to the formula H2?V G=eV G1V ET,where V G is the among-genotype variance compo-nent and V E is the residual(error)variance component.
Genetic correlations(r G)were estimated as r G?cov1;2=?????????????????????
V G13V G2
p
,where cov1,2is the covariance of trait means and V G1and V G2are the among-genotype variance components for those traits.The coef?cient of genetic variation(CV G)was estimated for each trait as CV G?e1003
??????
V G
p
T=X,where V G is the among-genotype variance component and X is the trait mean of the genotypes.
QTL analyses in the RIL population:To map QTL using the RIL population,a set of144markers equally spaced over the Arabidopsis genetic map was selected from the RIL L er/Cvi map(A lonso-B lanco et al.1998a).These markers spanned 485cM,with an average distance between consecutive markers of3.5cM and the largest genetic distance being11cM.The phenotypic values recorded,except basal branch number, were transformed(log10(x11))to improve the normality of the distributions and the values of16plants per RIL were used to calculate the means of each line for all traits.These means were used to perform the QTL analyses unless otherwise stated.The computer program MapQTL version5.0(V an O oijen2004)was used to identify and locate QTL linked to the molecular markers,using both interval mapping and multiple QTL mapping(MQM).In a?rst step,putative QTL were identi?ed using interval mapping.Thereafter,a marker closely linked to each putative QTL was selected as a cofactor and the selected markers were used as genetic background controls in the approximate MQM of MapQTL.LOD statistics were calculated at0.5-cM intervals.Tests of1000permutations were used to obtain an estimate of the number of type1errors (false positives).The genomewide LOD score,which95%of the permutations did not exceed,ranged from2.6to2.8and chromosomewide LOD thresholds varied between1.8and2.1 depending on trait and linkage group.The genomewide LOD score was then used as the signi?cance threshold to declare the presence of a QTL in MQM mapping,while the chromo-somewide thresholds were used to detect putative small-effect QTL.In the?nal MQM model the genetic effect(m Bàm A) and percentage of explained variance were estimated for each QTL and2-LOD support intervals were established as an $95%con?dence level(V an O oijen1992),using restricted MQM mapping.
Epistatic interactions between QTL were estimated using factorial analysis of variance.For each trait,the mean phe-notypic values were used as a dependent variable and co-factors,corresponding to the detected QTL,were used as?xed factors.The general linear model module of the statistical package SPSS version11.0was used to perform a full factorial analysis of variance or analysis of main effects only.Differences in R2-values,calculated from the type III sum of squares,were assigned to epistatic interaction effects of detected QTL.Addi-tionally we performed a complete pairwise search(P,0.001, determined by Monte Carlo simulations)for conditional and
Comparing a RIL and a New NIL Mapping Population893
coadaptive epistatic interactions for each trait,using the computer program EPISTAT(C hase et al.1997).
The effect of replication on statistical power was analyzed by performing MQM mapping on means of trait values from1,2, 4,8,12,and16replicate plants,respectively.Analyses were performed on10independent,stochastically sampled,data sets for each replication size and trait using automated cofactor selection(P,0.02).Total explained variance,LOD score of the largest-effect QTL,and number of signi?cant QTL were recorded for each analysis.
The effect of population size on statistical power was analyzed by performing MQM mapping on increasing pop-ulation sizes.Analyses were performed on10independent, stochastically sampled,data sets for each population size. Subpopulations of increasing size,with a step size of20lines, were analyzed for each trait using automated cofactor selec-tion(P,0.02).Total explained variance,LOD score of the largest-effect QTL,and number of signi?cant QTL were recorded for each analysis.
Statistical analyses of NILs:Differences in mean trait values of L er and NILs were analyzed by univariate analysis of variance,using the general linear model module of the sta-tistical package SPSS version11.0.Dunnett’s pairwise multiple comparison t-test was used as a post hoc test to determine signi?cant differences.For each analysis,trait values were used as a dependent variable and NILs were used as a?xed factor. Tests were performed two sided with a Bonferroni-corrected signi?cance threshold level of0.05and L er as a control category.To increase statistical power,similar analyses were conducted for bins(see results).For this,trait values of all introgression lines assigned to a certain bin were pooled and compared to values of the L er parental line.Because each NIL can be a member of more than one bin the signi?cance threshold was lowered to0.001to correct for multiple testing. The genetic effect of Cvi bins signi?cantly differing from L er was calculated as m Bàm A,where m A and m B are the mean trait values of L er and the Cvi bin,respectively.Explained variance was estimated from the partial h2of the univariate analysis of variance,where h2is the proportion of total variance attribut-able to factors in the analysis.The total percentage of explained variance was then estimated by using trait values as a dependent variable and NILs as a?xed factor,where all NILs were included as subjects.The percentage of explained variance of individual QTL was estimated as a fraction of the total variation in the population(including all lines),using a single bin as a?xed factor and as a fraction of the total var-iation in a comparison of a single bin with L er only.
To determine the effect of replicated measurements we calculated the power of detecting signi?cant differences between L er and NILs using various replicate numbers.For each trait we calculated the minimal relative difference in mean trait values that could still be signi?cantly detected. Calculations were performed using a normal distribution two-sample equal variance power calculator from the UCLA department of statistics(https://www.docsj.com/doc/477639813.html,/). We?rst calculated for each trait the mean phenotypic value of96L er replicate plants(m A)and for each line the standard deviation of24replicate plants.The mean line standard deviation of each trait was taken as a measure of variation (s)in all subsequent calculations.The signi?cance level,the probability of falsely rejecting the null hypothesis(H0:m A?m B)when it is true,was set to0.05and power,the probability of correctly rejecting the null hypothesis when the alternative (H1:m A?m B)is true,was set to0.95.The sample size of L er (N A)was always identical to the sample size of NILs(N B)and ranged from2to24individuals.For each trait and sample size the mean trait value(m B)for NILs was then calculated as the minimum value to meet the alternative hypothesis(H1:m A?m B)in a two-sided test.These minimum values were then converted in a fold-difference value compared to the L er value, calculated as(j m Bàm A j1m A)/m A,to obtain a relative estimate independent of trait measurement units.
The effect of replication on statistical power was also analyzed by performing bin mapping using2,4,8,12,and 16replicate plants,respectively.Analyses were performed on 10independent,stochastically sampled,data sets for each replication size and trait and the number of signi?cant QTL was recorded for each analysis.
RESULTS
Construction of a genomewide near-isogenic line population:We constructed a population of92in-trogression lines carrying between one and four Cvi introgression fragments in a L er genetic background. Lines were genotyped using349AFLP and95PCR markers to determine the number,position,and size of the introgressions(see materials and methods).This set of lines was selected to provide together an almost complete genomewide coverage(Figure1).Forty lines contained a single introgression while52lines carried several Cvi fragments.From those,32,19,and1line bore two,three,and four introgressions,respectively. The genetic length of the introgression fragments was estimated using the map positions of the introgressed markers in the genetic map constructed from the existing RIL population derived from the same L er and Cvi parental accessions(A lonso-B lanco et al. 1998a).The average genetic sizes of the main,second, third,and fourth introgression fragments were31.7, 11.1,6.7,and5.2cM,respectively.Thus,lines with multiple Cvi fragments carried a main large introgres-sion and several much smaller Cvi fragments.Addition-ally,we selected a core set of25lines that together covered.90%of the genome(supplemental Table1at https://www.docsj.com/doc/477639813.html,/supplemental/).
Genetic analyses of developmental traits:Six traits were measured and analyzed in the RIL and NIL populations(Table1).Although plants were grown in four replicated blocks,block effects were negligible and were therefore not used as a factor in subsequent analyses.In both populations,among-genotype vari-ance was highly signi?cant(P,0.0001)for all traits.In the RIL population,broad sense heritability estimates ranged from0.34(basal branch number)to0.92(total plant length)(Table1).Statistical parameters of most traits were similar to those described by A lonso-B lanco et al.(1998b,1999)and J uenger et al.(2005b).However, U ngerer et al.(2002)reported much lower average values for plant height and branch number although time to?ower was similar.Moreover,among-genotype variance estimates were lower and within-genotype var-iance estimates higher,resulting in lower heritability values compared to our analyses.
For the NIL population,mean trait values were closer to those measured for L er due to the genetic structure of
894J.J.B.Keurentjes et al.
F igure 1.—Graphical genotype of the L er /Cvi NIL population.Bars represent introgressions.Solid bars represent the genetic position of Cvi introgressions in individual NILs.Shaded bars represent crossover regions between markers used for the geno-typing of the lines.Numbers at the top indicate the ?ve linkage groups.
Comparing a RIL and a New NIL Mapping Population 895
the population,consisting of lines carrying only small Cvi introgressions in a L er background.Furthermore,variance components from ANOVA were lower in the NIL population but heritability estimates differed only slightly compared to the RIL population (Table 1).Strong and similar genetic correlations were observed between traits in the two L er /Cvi populations,indicat-ing partial genetic coregulation (Table 2).Flowering time shows the highest correlation with the number of main in?orescence branches but is negatively corre-lated with basal branch number.Flowering time is also,but to a lesser degree,correlated with plant height.Correlations were also found between plant height and
branching with again positive values with the number of main in?orescence branches and negative correlations with basal branch number.These results contrasted with those from U ngerer et al.(2002),who found negative correlations between ?owering time,plant height,and branching in all pairwise comparisons,which is proba-bly due to the different environmental setups in the two laboratories.
Mapping quantitative traits in the L er /Cvi RIL pop-ulation:Each trait was subjected to QTL analysis and three to eight QTL were detected for each trait (Figure 2,Table 3).Major QTL for ?owering time,plant height,and branching were in concordance with pre-viously reported studies (A lonso -B lanco et al.1998b,1999;U ngerer et al.2002,2003;J uenger et al.2005b),although slight differences for minor QTL were also found.Total explained variance for each trait ranged from 38.5%for basal branch number to 86.3%for total plant height.LOD scores for the largest-effect QTL ranged from 5.7for basal branch number up to 60.7for total plant height with corresponding explained vari-ances of 11.0and 64.0%,respectively.The average ge-netic length of 2-LOD support intervals was 11.6cM,ranging from 2.3(length at ?rst silique)to 33.3cM (total branch number).Opposing-effect QTL were found for all traits,explaining the observed transgressive segre-gation within the population (data not shown).Genetic interaction among the detected QTL was also tested.The proportion of variance explained by epistatic in-teractions ranged from 3.1(basal branch number)to 20.5%(number of main in?orescence branches)and involved two to ?ve of the detected QTL (Table 3).Using a complete pairwise search of all markers (C hase et al.1997),a number of additional interactions were de-tected between loci not colocating with major QTL po-sitions (supplemental Figure 1at https://www.docsj.com/doc/477639813.html,/supplemental/).
The smallest signi?cant absolute effect detected was 4.4days for ?owering time,1.0and 2.3cm for length at
TABLE 1
Descriptive statistics for six developmental traits analyzed in
two mapping populations and their parents Trait X 6eSD T[V G ]a [V E ]b [H 2]c [CV G ]d Parents
FT (days)24.30(1.03)e
8.74
3.570.7110.8530.21(2.47)f
SL (cm)9.58(0.98)e 3.27
3.140.5115.8713.21(2.30)
f
TL (cm)23.59(1.92)e 26.8110.530.7217.9933.95(4.17)f IB 2.21(0.46)e 0.02
0.330.05 5.532.49(0.67)f
BB 1.54(0.68)e 0.00
0.650.000.001.48(0.91)f
TB
3.75(0.77)e 0.01
0.82
0.01
1.88
3.97(1.02)
f
RIL population FT (days)26.06(6.03)32.59 3.820.90
21.91SL (cm)9.89(3.39)9.70 1.800.8331.49TL (cm)26.13(9.22)78.53 6.520.9233.91IB 2.34(1.22)0.990.500.6742.66BB 1.43(0.93)0.300.570.3437.98TB 3.77(1.27)0.780.840.4823.36NIL population FT (days)23.68(3.60)10.78 2.210.8313.87SL (cm)9.81(2.18) 3.17 1.580.6518.15TL (cm)24.50(5.95)31.24 4.100.8722.82IB 2.26(0.88)0.510.270.6531.42BB 1.56(0.84)0.180.530.2426.92TB
3.82
(1.06)0.480.64
0.42
18.25
FT,?owering time;SL,length until ?rst silique;TL,total plant length;IB,main in?orescence branch number;BB,basal branch number;TB,total branch number.a
Among-genotype variance component from ANOVA:tests whether genetic differences exist among genotypes for spec-i?ed traits (P ,0.0001).b
Residual variance component from ANOVA.c
Measure of total phenotypic variance attributable to ge-netic differences among genotypes (broad sense heritability)calculated as V G /(V G 1V E ).d
Coef?cient of genetic variation calculated as e1003??????V G p T=X .e
Landsberg erecta parent.f
Cape Verde Islands parent.
TABLE 2
Genetic correlations among developmental traits analyzed in
two mapping populations Trait FT SL TL IB BB TB FT 0.63*0.38*0.97*à0.49*0.80*SL 0.39*0.90*0.52*à0.39*0.35*TL 0.21*0.88*0.18*à0.32*0.00IB 0.91*0.31*0.09*à0.54*0.95*BB à0.26*à0.28*à0.26*à0.35*0.12*
TB
0.77*
0.15*
à0.07
0.85*
0.31*
The top right and the bottom left halves of the table repre-sent values calculated for the RIL and the NIL populations,respectively.FT,?owering time;SL,length until ?rst silique;TL,total plant length;IB,main in?orescence branch num-ber;BB,basal branch number;TB,total branch number.*Signi?cant at P ,0.001.
896
J.J.B.Keurentjes et al .
?rst silique and total plant length,respectively,and 0.3,0.3,and 0.4for the number of main in?orescence branches,basal branch number,and total branch num-ber,respectively.Relative effects,expressed as the fold difference between genotypes,calculated as (j m B àm A j 1m A )/m A ,then equaled 1.15-,1.09-,1.09-,1.13-,1.59-,and 1.10-fold,respectively (Tables 3and 5).As expected,the total explained variance of a trait correlated positively with the smallest signi?cantly detectable effect for that particular trait.In general,smaller effects could be de-tected with increasing total explained variance.When the chromosomewide threshold for signi?cance was used instead of the genomewide threshold,one addi-tional suggestive QTL was detected for main in?ores-cence branch number and total branch number and two for length at ?rst silique.
Mapping quantitative traits in the L er /Cvi NIL population:To search for QTL in the NIL population,we divided the Arabidopsis genetic map in adjacent genomic fragments that were individually tested.The complete genome was subdivided into 97regions,de?ned by the position of the recombination events of the main introgressions of the 92NILs (supplemental Table 2at https://www.docsj.com/doc/477639813.html,/supplemental/).These regions are referred to as bins and each NIL was then assigned to those adjacent bins spanned by its Cvi introgression fragment.Thus,each bin contains a unique subset of lines with overlapping Cvi introgres-sions in that particular region,which were used to test the phenotypic effects of that bin.The average genetic length of the bins was 5.0cM,ranging from 0.1to 26.3cM.The number of NILs per bin ranged from 0to 13with an average of 5.1NILs.Because NILs were assigned only to bins when the complete bin was covered by the introgression,3bins remained empty [viz.bins 66(26.3cM),73(3.3cM),and 77(5.4cM)].On average each NIL was assigned to 5.4adjacent bins.One NIL (LCN4-2)was not assigned to any bin because its introgression included only a single marker.Two NILs corresponded to complete chromosomal substitutions:line LCN3-8(chromosome 3)and line LCN1-8(chromosome 1),the latter carrying the largest introgression assigned to 27adjacent bins.
To map QTL in the NIL population,all bins were tested individually by comparing the phenotypes of the NILs assigned to each bin with that of L er .As shown in Figure 3and Table 4,one to nine QTL were detected for each trait.The total explained variance for each trait ranged from 26.7%for basal branch number up to 87.7%for total plant height.Explained variances for the largest-effect QTL for each trait ranged from 19.3%for basal branch number to 91.9%for total plant height as calculated from a restricted ANOVA using only lines from the most signi?cant bin and L er .To show the relative effect of Mendelizing QTL with respect to the total population variance we calculated the explained variances also when all lines of the population were subjected to ANOVA analysis using the most signi?cant bin as a ?xed factor (Table 4).Relative effects of QTL were much lower in this unrestricted analysis because all other QTL in the population increase residual variation that is not corrected for,as is done in MQM mapping in the RIL population.Moreover,lines partly overlapping the QTL bin are not assigned to that bin but can still contain the QTL Cvi allele,further increasing the residual variation in the population.
The smallest signi?cant QTL effect detected was 0.7days for ?owering time,1.1and 2.1cm for length at ?rst silique and total plant length,respectively,and 3.8,0.5,and 0.4for the number of main in?orescence branches,basal branch number,and total branch number,respec-tively.Relative effects,expressed as the fold
difference
F igure 2.—Genomewide QTL pro?les of traits analyzed in the RIL population:(A)?owering time,(B)length at ?rst silique,(C)total plant length,(D)number of main in?orescence branches,(E)basal branch number,and (F)total branch number.Solid lines represent the QTL effect calculated as described in materials and methods .Shaded lines represent LOD scores.Shaded dashed lines represent genomewide signi?cance threshold levels for LOD scores determined by permutation testing.
Comparing a RIL and a New NIL Mapping Population 897
between genotypes,calculated as(j m Bàm A j1m A)/m A, then equaled1.03-,1.11-,1.09-,2.71-,1.30-,and1.11-fold,respectively(Tables4and5).
For a number of traits several QTL were found that could not be signi?cantly detected in the RIL popula-tion.In total12of such small-effect QTL were detected for?owering time(3),length at?rst silique(5),total plant length(2),and basal branch number(2).None of those met the lower chromosomewide signi?cance threshold for suggestive QTL in the RIL population.Although2were close to this threshold,10of them did not reach LOD scores.1.0in the RIL popula-tion(supplemental Table3at https://www.docsj.com/doc/477639813.html,/ supplemental/).
We de?ned the support interval in the NIL mapping population as the region spanned by consecutive bins, signi?cantly differing from L er(P,0.001)and sharing the same direction of effect.The length of support intervals estimated in this way ranged from1.4(total plant length)to85.3cM(basal branch number)with an
TABLE3
QTL detected in the RIL population
Trait Chr a LOD
score
Support
interval b(cM)
Explained
variance c(%)Effect d
Total explained
variance e(%)Interaction f(%)
FT111.9 1.5–9.8g13.0à3.968.49.6 518.9388.4–394.5g22.2 5.7
511.9408.2–413.7g13.0 4.4
SL19.30.0–9.3 6.3à1.779.515.0
1 4.8103.1–126.0 3.1à1.3
239.7173.2–175.543.2 4.5
3 2.9234.2–253.6 1.9 1.0
3 5.0281.5–287.8 3.2à1.2
515.7387.9–392.4g11.8 2.9
510.2403.6–409.7g7.2 2.0
TL1 6.50.0–9.8g 2.8à3.186.311.5
1 5.073.9–84.6 2.1à2.7
1 3.3116.3–126.0 1.2à2.3
260.7173.2–176.0g64.014.8
3 6.0207.3–225.7g 2.6à3.0
4 5.2287.8–307.5g 2.2à2.7
57.8383.1–392.5g 3.6 4.1
5 5.1403.6–411.7 2.2 3.0
IB1 5.00.0–13.5g 5.3à0.465.020.5
2 2.7154.9–171.0g 2.8à0.3
515.3387.0–391.9g19.70.9
510.4398.8–411.7g12.30.7
5 3.1472.2–485.3 3.2à0.3
BB1 5.772.4–91.0g11.00.438.5 3.1
2 3.2167.0–200.2g 6.2à0.3
4 4.6360.7–373.5g9.10.4
5 5.5385.6–406.1g11.3à0.5
TB115.5 5.3–12.4g16.1à0.871.116.2
1 4.981.7–93.8g 4.60.4
29.5169.0–180.0g9.1à0.6
59.7386.5–392.4g9.40.6
510.9403.3–412.2g10.80.7
5 5.2472.2–485.3 4.7à0.4
FT,?owering time;SL,length until?rst silique;TL,total plant length;IB,main in?orescence branch num-ber;BB,basal branch number;TB,total branch number.
a Chromosome number.
b2-LOD support interval.
c Percentage of total variation explaine
d by individual QTL.
d Effect of QTL calculated as m Bàm A,wher
e A and B are RILs carrying L er and Cvi genotypes at the QTL
position,respectively.m B and m A were estimated by MapQTL.Effects are given in days(?owering time),cen-timeters(length at?rst silique and total length),or numbers(elongated axils,basal branch number,and total branch number).
e Percentage o
f total variance explained by genetic factors estimated by MapQTL.
f Percentage of total variation explained by interaction between individual QTL.
g QTL showing signi?cant epistatic interactions(P,0.05)and used to estimate the percentage of explained
variance by genetic interactions.
898J.J.B.Keurentjes et al.
average of 30.9cM.Alternatively,we also searched for QTL in the NIL population by comparing the pheno-type of each NIL individually against L er (supplemental Figures 2–7at https://www.docsj.com/doc/477639813.html,/supplemental/).In this case,support intervals can be estimated as the length of the overlapping regions between the Cvi introgression fragments of NILs signi?cantly differing from L er in a particular genomic region.This second method increases the QTL localization resolution,but reduces statistical power.For each bin on average 116plants could be tested against L er whereas only 24plants were available for analysis of individual NILs.Moreover,individual lines may contain multiple opposing-effect QTL,resulting in nonsigni?cant differences compared to L er .Therefore,lines spanning the bin support inter-val were occasionally not signi?cantly different from L er .Likewise,lines bearing introgressions outside the bin support intervals sometimes differed signi?cantly from L er ,probably due to multiple additive small-effect QTL.Together,the loss of power and the complexity of the traits under study hindered a con?dent estimation of a NIL support interval.Nevertheless,all QTL detected in the bin analysis could also be detected by analyzing individual NILs.As a compromise between the two me-thods of support interval estimation we recorded the position of the largest-effect bin within the bin support interval (Table 4).However,it must be noted that bin support intervals may contain multiple QTL of similar direction.The average size of these largest-effect bins was 4.6cM.Within those bins,at least one individual NIL signi?cantly differing from L er was always found.Power in RIL vs.NIL QTL mapping:The power to detect a QTL at a speci?c locus basically depends on the difference in mean trait values between A and B genotypes for that particular locus.Although other parameters like trait heritability,genetic interactions,and genetic map quality should not be ignored.Because power increases when variance for mean values de-creases,QTL analyses can bene?t greatly from multiple measurements.In a RIL population this can be achieved in two ways.First,because segregation of both alleles occurs randomly and each locus is represented equally by the A and the B genotype,provided there is no segregation distortion (D oerge 2002),increasing the number of RILs to be analyzed will increase the number of observations of each genotype at a given genomic position.A further advantage of increasing the RIL population size is that the number of recombination events increases,which can improve resolution.How-ever,when the number of lines is ?xed,more accurate trait values of lines can be achieved by measuring rep-licate individuals of the same line.In addition,accurate trait values based on replicate measurements improve the possibility of detecting smaller-effect QTL.
To test the effect of replicated measurements and population size on the QTL detection power of the two L er /Cvi populations we analyzed the phenotypic data obtained in these populations by varying both parame-ters.For the RIL population we performed QTL ana-lyses on different numbers of RILs (population size)and used mean line values obtained with different numbers of replicates (replicate size).The total ex-plained variance in the population,the LOD score of the largest-effect QTL,and the number of detected QTL were then recorded for each trait (Figure 4).When the population size was kept constant (161lines),the recorded statistics increased when increasing the repli-cate number from one to four but this increase
leveled
F igure 3.—QTL pro?les of traits analyzed in the NIL population:(A)?owering time,(B)length at ?rst silique,(C)total plant length,(D)number of main in?orescence branches,(E)basal branch number,and (F)total branch number.Solid lines represent the QTL effect calculated as described in materials and methods .Shaded lines represent signi?cance scores.Shaded dashed lines represent signi?cance threshold levels applied in this study.
Comparing a RIL and a New NIL Mapping Population 899
off rapidly when measuring more replicates(Figure4, A–C).In contrast,when the number of replicates was kept constant(16replicated measurements per RIL) and population size was increased,the QTL detection power improved more drastically.However,the total explained variance remained more or less constant over all population sizes(Figure4D).This phenomenon is commonly known as the Beavis effect and is due to the fact that estimated explained variances of detected QTL are sampled from a truncated distribution because QTL are taken into account only when the test statistics reach a predetermined critical value(X u2003).As a result, the expectations of detected QTL effects are biased upward.A second effect of increasing population size is the nearly linear increase of LOD scores,observed for all analyzed QTL(Figure4E).Signi?cance thresholds determined by permutation tests for each population size were steady around2.7LOD for population sizes
TABLE4
QTL detected in the NIL population
Support interval b(cM)
Support
bin(cM)c
Explained variance(%)
Effect f
Total explained
variance g(%)
Trait Chr a Restricted d Unrestricted e
FT10.0–21.6 3.9–7.870.3 3.2à3.283.2 131.4–40.633.4–40.718.00.5à1.0
173.3–122.083.6–877.10.7à0.7
2174.4–204.7200.9–201.822.30.6 1.5
5388.4–434.2392.3–39552.142.815.7
SL110.8–27.417.3–21.764.0 4.8à3.166.1 131.4–40.633.4–40.717.10.6à1.1
173.3–125.9122.1–12634.9 2.8à1.7
2160.8–207.2162–174.573.4 5.3 4.9
3270.1–288.4287.1–288.437.1 1.6à1.7
4359.5–375.7368.2–375.732.2 1.7à1.6
5388.3–418.9392.3–39532.20.7 2.7
5434.2–436.0434.3–436.129.6 3.8à1.4
5441.4–459.3454.3–459.428.2 1.1à1.1
TL10.0–33.317.3–21.766.2 1.7à6.387.7 164.7–125.9122.1–12648.8 3.8à3.8
2160.8–207.2174.5–178.891.910.518.5
3287.0–288.4287.1–288.419.00.4à2.1
5389.9–416.1411.7–416.234.1 1.7 3.7
5434.2–454.3434.3–436.145.0 1.4à3.9
IB5388.3–434.2392.3–39546.337.7 3.866.1
BB10.0–15.1 3.9–7.817.7 1.8à0.626.7 140.6–125.994.5–101.617.99.00.8
2174.4–189.1179.7–189.211.4 2.4à0.5
5388.3–434.2392.3–39514.4 1.7à0.7
5483.2–487.8483.2–487.819.3 1.1à0.8
TB10.0–15.97.8–9.924.1 2.2à0.844.1 140.6–125.994.5–101.614.0 4.10.8
2174.4–189.1179.7–189.27.6 1.5à0.4
5388.3–434.2392.3–39543.217.4 3.1
FT,?owering time;SL,length until?rst silique;TL,total plant length;IB,main in?orescence branch num-ber;BB,basal branch number;TB,total branch number.
a Chromosome number.
b The region spanned by consecutive bins,signi?cantly(P,0.001)differing from L er and sharing the same
direction of effect,was taken as the support interval.
c Position of the bin within the QTL support interval showing the largest effect.
d Within th
e QTL support interval the bin showing the largest effect was compared to L er in an ANOVA anal-
ysis.The among-genotype component of ANOVA was taken as an estimator of explained variance.
e All lines in the population were subjected to ANOVA using the bin described in footnote d as a?xed factor.
The among-genotype component of ANOVA was taken as an estimator of explained variance.
f Effect of QTL calculated as m Bàm A,where m A is the mean value of all L er lines and m B is the mean value of all
lines in the bin described in footnote d.Effects are given in days(?owering time),centimeters(length at?rst silique and total length),or numbers(main in?orescence branch number,basal branch number,and total branch number).
g All bins together with L er were analyzed by ANOVA and the among-genotype component was taken as a
measure of totally explained variance.
900J.J.B.Keurentjes et al.
.30RILs and increased slightly with smaller population sizes(data not shown).The largest-effect QTL could be signi?cantly detected at all population sizes for all traits except for basal branch number,whose largest-effect QTL could not be signi?cantly detected in population sizes,80RILs.
To evaluate the NIL population,we studied the effect of increasing the number of replicates per line by esti-mating the relative difference between line mean values that could still be signi?cantly detected with different replicate numbers(see materials and methods).As shown in Figure5A the power to detect signi?cant phe-notypic differences greatly increases when increasing the number of replicate individuals of NILs measured. Furthermore,the lower the heritability of the trait the larger the increase of detection power achieved by in-creasing the number of replicates per NIL.When a bin analysis was carried out using increasing replicate num-bers a similar increase in the number of detected QTL was observed(Figure5B).Overall,the results presented in Figures4and5show that the number of replicates used in our analyses(16individuals for each RIL and24 individuals for each NIL)approximated the maximum QTL detection power of both L er/Cvi populations.
DISCUSSION
Experimental mapping populations are a basic re-source to elucidate the genetic basis of quantitative multigenic traits.In this work,we have developed the ?rst genomewide population of NILs of Arabidopsis thaliana consisting of92lines carrying genomic intro-gression fragments from the parental accession Cvi into the common laboratory genetic background Landsberg erecta.In addition we have empirically compared the mapping power of this population with that of an ex-isting population of recombinant inbred lines,derived from the same parental accessions.RIL and NIL pop-ulations have been used extensively in genetic studies (E shed and Z amir1995;R ae et al.1999;M onforte and T anksley2000;K oumproglou et al.2002;H an et al. 2004;K oornneef et al.2004;S inger et al.2004;V on K orff et al.2004)due to the advantages derived from their homozygosity and immortality:they can be used inde?nitely;various traits can be analyzed in different experiments and environmental settings;and replicates of the individual lines can be analyzed,enabling a more accurate estimate of the line’s phenotypic mean value. However,the main difference between the two popula-tions lies in the nature of their genetic makeup.In a RIL population multiple genomic regions differ between most pairs of RILs and several segregating QTL con-tribute to phenotypic differences between pairs of lines, making it impossible to assign the observed variation between pairs of lines to a speci?c genomic region. Therefore,to detect QTL one must perform the simul-taneous analysis of a large number of lines.In con-trast,in a NIL population,the phenotypic variation observed between pairs of lines can be assigned directly to the distinct genomic regions introgressed in an
TABLE5
Comparative summary of QTL mapping parameters in the L er/Cvi RIL and NIL populations
Trait Population a QTL b
(no.)
Support c
(cM)
Explained
variance d(%)
Total explained
variance(%)Effect e
Relative
effect f
FT RIL3 6.616.168.4 4.7 1.15 NIL535.5(3.6)34.083.2 4.4 1.03
SL RIL710.111.079.5 2.1 1.09 NIL923.3(5.2)38.766.1 2.1 1.11
TL RIL811.110.186.3 4.5 1.09 NIL631.4(3.4)50.887.7 6.4 1.09
IB RIL512.18.765.00.5 1.13 NIL145.9(2.7)46.366.1 3.8 2.71
BB RIL421.39.438.50.4 1.59 NIL533.1(5.6)16.126.70.7 1.30
TB RIL69.79.171.10.6 1.10 NIL440.5(5.4)22.244.1 1.3 1.11
FT,?owering time;SL,length until?rst silique;TL,total plant length;IB,main in?orescence branch num-
ber;BB,basal branch number;TB,total branch number.
a Population type.
b Number of QTL detected.
c Average length of support interval.In parentheses:average length of largest-effect bin.
d Averag
e explained variance for each QTL.
e Average absolute effect for each QTL.Effects are given in days(?owering time),centimeters(length at?rst
silique and total length),or numbers(elongated axils,basal branch number,and total branch number).
f Smallest relative effect signi?cantly detected,expressed as fold difference compared to L er,calculated as
(j m Bàm A j1m A)/m A.
Comparing a RIL and a New NIL Mapping Population901
otherwise similar genetic background.Depending on the desired resolution one can minimize the number of lines by analyzing lines carrying large introgressions or even chromosome substitution strains (N adeau et al.2000).
A summary of the differences observed between the RIL and NIL populations derived from L er and Cvi is shown in Table 5and in supplemental Figure 8at https://www.docsj.com/doc/477639813.html,/supplemental/.The total number of QTL detected did not differ much between the two populations.However,different loci were detected in both types of populations,showing their complemen-tary properties.For both populations the detection of QTL was highly dependent on the trait under consid-eration and its genetic architecture (e.g.,effect and position of QTL,epistasis).The power of the new NIL population to detect the large-effect loci was close to that of the existing RIL population since most large-effect loci were detected in both populations.However,a few relatively large-effect loci showing signi?cant epi-static interactions could be detected only in the RIL population,but not in the NILs (supplemental Table 3at https://www.docsj.com/doc/477639813.html,/supplemental/).Moreover,localization resolution was higher in the RIL population
compared to the bin analysis of the NIL population,allowing separation of linked QTL.This was best illus-trated by the two major QTL for ?owering time detected in the RIL population on the top of chromosome 5,which not only are linked but also showed strong epi-static interaction.Consequently,these two QTL could not be separated in the NIL population.Nevertheless,the QTL resolution in the NIL population can be in-creased when analyzing individual lines,although this will be at the cost of mapping power.In total,14QTL detected in the RIL population could not be detected in the NIL population,of which 10showed signi?cant epistatic interaction with other QTL and all others were closely linked to another signi?cant QTL.
In contrast,the average explained variance of single QTL was higher in the NIL population,increasing the power to detect small-effect QTL.This difference can be attributed to the level of transgression,which is stronger in the RIL population,thereby increasing total phenotypic variance.As a result,13small-effect QTL could be detected in the NIL population,which were not detected in the RIL population.Nevertheless,
some
F igure 4.—QTL detection power analysis of the L er /Cvi RIL population.(A)Effect of replicate number on total explained variance.(B)Effect of replicate number on LOD score of the largest-effect QTL.(C)Effect of replicate number on the number of detected QTL.(D)Effect of population size on total explained variance.(E)Effect of population size on LOD score of the largest-effect QTL.(F)Effect of population size on the number of detected QTL.h ,?owering time;e ,length at ?rst silique;D ,total plant length;x ,main in?orescence branch number;s ,basal branch number;and 1,total branch number.Error bars rep-resent SEM of 10independent analyses.
902J.J.B.Keurentjes et al .
of the small-effect QTL detected in the NILs were close to the signi?cance threshold in the RIL population when using the lower chromosomal LOD thresholds (supplemental Table3at https://www.docsj.com/doc/477639813.html,/ supplemental/).Expectedly,the power to detect small-effect QTL in the NIL population was higher for the more heritable traits(?owering time and plant height) than for those traits with low heritability(branching traits).The different power to detect small-effect QTL of the two populations is due to the effect of the segrega-tion of multiple QTL in the RIL population,which in-creases the residual variance at each QTL under study. The analyses of the RIL and NIL populations per-formed in this work were probably close to the maxi-mum statistical power for the given population sizes since the number of detected QTL leveled off at higher replication sizes(Figures4and5).The power analyses presented here could guide the making of decisions on the number of plants to be analyzed when experiments are costly,laborious,or time consuming and therefore may require the analysis of fewer plants.Overall,for RILs,the effect of population size on mapping power was larger than the effect of replicated measurements of individual lines.Therefore,to reduce the number of plants to be analyzed,it is preferable to?rst reduce the number of replicates per line,and only thereafter,if required,the number of lines.In our analyses major-effect QTL for most traits could still be signi?cantly detected when only50lines were analyzed without replicates(data not shown).However,due to the Beavis effect(X u2003)the explained variances obtained with small population sizes were strongly overestimated.In the NIL population,the number of replicated measure-ments has a larger impact on mapping power and at least?ve replicated plants should be analyzed to obtain enough statistical power(Figure5).However,fewer lines can be analyzed as long as genomewide coverage is maintained.In this NIL population this can be achieved using a core set of25lines,although localization res-olution was diminished.Nevertheless,most QTL de-tected in the full set could still be detected in the core set(supplemental Figure9at https://www.docsj.com/doc/477639813.html,/ supplemental/).Once a QTL has been identi?ed in a particular region,one can zoom in with a minimal set of lines carrying smaller introgressions de?ned by cross-overs in the support interval of the QTL of interest (F ridman et al.2004).
The L er/Cvi NIL population developed in this work provides a useful resource that will facilitate the genetic dissection of quantitative traits in Arabidopsis in various aspects.First,as shown here,it can be analyzed as an alternative segregating population to perform genome-wide QTL mapping,with the particular advantage of detecting small-effect QTL.Second,this population can be used to con?rm previously detected QTL in the L er/ Cvi RIL population.Third,individual lines of this pop-ulation can serve as a starting point for the rapid Mendelization of particular QTL and for their?ne mapping and cloning(P aran and Z amir2003).Finally, the single introgression lines of this population may also strongly facilitate the?ne mapping of arti?cially in-duced mutant alleles in the common laboratory L er genetic background(or transferred to this accession). The?ne mapping of mutant loci affecting quantitative adaptive traits is often hampered by the confounding effects of QTL segregating in the mapping populations derived from crosses between the mutant and another Arabidopsis wild accession.Knowing the approximate genetic location of the mutant locus within a chromo-somal arm,speci?c lines of this NIL population can be selected as carrying a single introgression spanning the map position of the locus of interest.These lines can then be used to derive the required monogenic mapping population,as has been illustrated with the ?owering-time locus FVE(A usin et al.2004).In conclu-sion,the elucidation of quantitative traits can bene?t from the parallel analysis of both populations.
We thank Kieron Edwards for sharing NILs,Johan van Ooijen for helpful assistance in the QTL mapping,and Piet Stam for critical reading of the manuscript.This work was supported by a grant from The Netherlands Organization for Scienti?c Research,Program Genomics
(050-10-029).
F igure 5.—QTL detection power analysis of the L er/Cvi NIL population.(A)Effect of replicate number on signi?cantly detectable relative differences,expressed as fold difference between two lines.
(B)Effect of replicate number on the number of detected QTL.h,?owering time;e,length at?rst si-lique;D,total plant length;x,main in?orescence branch number;s, basal branch number;and1,total branch number.Error bars rep-resent SEM of10independent analyses.
Comparing a RIL and a New NIL Mapping Population903
LITERATURE CITED
A lonso-
B lanco,C.,and M.K oornneef,2000Naturally occurring
variation in Arabidopsis:an underexploited resource for plant ge-netics.Trends Plant Sci.5:22–29.
A lonso-
B lanco,C.,A.J.M.P eeters,M.K oornneef,C.L ister,C.
D ean et al.,1998a Development of an AFLP based linkage
map of L er,Col and Cvi Arabidopsis thaliana ecotypes and con-struction of a L er/Cvi recombinant inbred line population.Plant J.14:259–271.
A lonso-
B lanco,C.,S.E.E l-A ssal,G.
C oupland and M.K oornneef,
1998b Analysis of natural allelic variation at?owering time loci in the Landsberg erecta and Cape Verde Islands ecotypes of Arabidopsis thaliana.Genetics149:749–764.
A lonso-
B lanco,C.,H.B lankestijn-D e V ries,C.J.H anhart and M.
K oornneef,1999Natural allelic variation at seed size loci in re-lation to other life history traits of Arabidopsis thaliana.Proc.Natl.
https://www.docsj.com/doc/477639813.html,A96:4710–4717.
A lonso-
B lanco,C.,L.B entsink,C.J.H anhart,H.B lankestijn-
D e V ries and M.K oornneef,2003Analysis of natural allelic
variation at seed dormancy loci of Arabidopsis thaliana.Genetics 164:711–729.
A usin,I.,C.A lonso-
B lanco,J.A.J arillo,L.R uiz-G arcia and
J.M.M artinez-Z apater,2004Regulation of?owering time by FVE,a retinoblastoma-associated protein.Nat.Genet.36: 162–166.
B entsink,L.,K.Y uan,M.K oornneef and D.V reugdenhil,
2003The genetics of phytate and phosphate accumulation in seeds and leaves of Arabidopsis thaliana,using natural variation.
Theor.Appl.Genet.106:1234–1243.
B lair,M.W.,G.I riarte and S.B eebe,2006QTL analysis of yield
traits in an advanced backcross population derived from a culti-vated Andean x wild common bean(Phaseolus vulgaris L.)cross.
Theor.Appl.Genet.112:1149–1163.
B lanco,A.,R.S imeone and A.G adaleta,2006Detection of QTLs
for grain protein content in durum wheat.Theor.Appl.Genet.
112:1195–1204.
B orevitz,J.O.,and M.N ordborg,2003The impact of genomics
on the study of natural variation in Arabidopsis.Plant Physiol.
132:718–725.
B roman,K.W.,2001Review of statistical methods for QTL mapping
in experimental https://www.docsj.com/doc/477639813.html,b Anim.30:44–52.
C hase,K.,F.R.A dler and K.G.L ark,1997Epistat:a computer
program for identifying and testing interactions between pairs of quantitative trait loci.Theor.Appl.Genet.94:724–730.
D oerge,R.W.,2002Mapping and analysis of quantitative trait loci
in experimental populations.Nat.Rev.Genet.3:43–52.
E dwards,K.D.,J.R.L ynn,P.G yula,F.N agy and A.J.M illar,
2005Natural allelic variation in the temperature-compensation mechanisms of the Arabidopsis thaliana circadian clock.Genetics 170:387–400.
E shed,Y.,and D.Z amir,1995An introgression line population of
Lycopersicon pennellii in the cultivated tomato enables the identi-?cation and?ne mapping of yield-associated QTL.Genetics141: 1147–1162.
F ridman, E., F.C arrari,Y.L iu, A.R.F ernie and D.Z amir,
2004Zooming in on a quantitative trait for tomato yield using interspeci?c introgressions.Science305:1786–1789.
H an,F.,S.E.U llrich,A.K leinhofs,B.L.J ones,P.M.H ayes et al.,
1997Fine structure mapping of the barley chromosome-1cen-tromere region containing malting-quality QTLs.Theor.Appl.
Genet.95:903–910.
H an,F.,J.A.C lancy,B.L.J ones,D.M.W esenberg,A.K leinhofs
et al.,2004Dissection of a malting quality QTL region on chro-mosome1(7H)of barley.Mol.Breed.14:339–347.
J ansen,R.C.,2003Quantitative trait loci in inbred lines,pp.445–476in Handbook of Statistical Genetics,edited by D.J.B alding, M.B ishop and C.C annings.John Wiley&Sons,Chichester, UK.
J euken,M.J.W.,and P.L indhout,2004The development of lettuce backcross inbred lines(BILs)for exploitation of the Lactuca saligna(wild lettuce)germplasm.Theor.Appl.Genet.
109:394–401.
J uenger,T.E.,J.K.M ckay,N.H ausmann,J.J.B.K eurentjes,S.S en et al.,2005a Identi?cation and characterization of QTL under-lying whole-plant physiology in Arabidopsis thaliana:d13C,stomatal
conductance and transpiration ef?ciency.Plant Cell Environ.28: 697–708.
J uenger,T.E.,S.S en,K.A.S towe and E.L.S imms,2005b Epistasis and genotype-environment interaction for quantitative trait loci affecting?owering time in Arabidopsis thaliana.Genetica123: 87–105.
K oornneef,M., C.A lonso-B lanco and D.V reugdenhil, 2004Naturally occurring genetic variation in Arabidopsis tha-liana.Annu.Rev.Plant Biol.55:141–172.
K oumproglou,R.,T.M.W ilkes,P.T ownson,X.Y.W ang,J.B eynon et al.,2002STAIRS:a new genetic resource for functional geno-mic studies of Arabidopsis.Plant J.31:355–364.
L oudet,O.,V.G audon, A.T rubuil and F.D aniel-V edele, 2005Quantitative trait loci controlling root growth and archi-tecture in Arabidopsis thaliana con?rmed by heterogeneous in-bred family.Theor.Appl.Genet.110:742–753.
M aloof,J.N.,2003Genomic approaches to analyzing natural var-iation in Arabidopsis thaliana.Curr.Opin.Genet.Dev.13:576–582.
M onforte,A.J.,and S.D.T anksley,2000Development of a set of near isogenic and backcross recombinant inbred lines contain-ing most of the Lycopersicon hirsutum genome in a L.esculentum genetic background:a tool for gene mapping and gene discovery.
Genome43:803–813.
N adeau,J.H.,J. B.S inger, A.M atin and E.S.L ander, 2000Analysing complex genetic traits with chromosome substi-tution strains.Nat.Genet.24:221–225.
P aran,I.,and D.Z amir,2003Quantitative traits in plants:beyond the QTL.Trends Genet.19:303–306.
R ae,A.M.,E.C.H owell and M.J.K earsey,1999More QTL for ?owering time revealed by substitution lines in Brassica oleracea.
Heredity83:586–596.
R eymond,M.,S.S vistoonoff,O.L oudet,L.N ussaume and T.
D esnos,2006Identi?cation of QTL controlling root growth
response to phosphate starvation in Arabidopsis thaliana.Plant Cell Environ.29:115–125.
S inger,J.B.,A.E.H ill,L.C.B urrage,K.R.O lszens,J.S ong et al., 2004Genetic dissection of complex traits with chromosome substitution strains of mice.Science304:445–448.
S late,J.,2005Quantitative trait locus mapping in natural popula-tions:progress,caveatsandfuturedirections.Mol.Ecol.14:363–379. S tylianou,I.M.,S.T saih,K.D ipetrillo,N.I shimori,R.L i et al., 2006Complex genetic architecture revealed by analysis of high-density lipoprotein cholesterol in chromosome substitution strains and F2crosses.Genetics174:999–1007.
S warup,K.,C.A lonso-B lanco,J.R.L ynn,S.D.M ichaels,R.M.
A masino et al.,1999Natural allelic variation identi?es new
genes in the Arabidopsis circadian system.Plant J.20:67–77.
T eng,S.,J.K eurentjes,L.B entsink,M.K oornneef and S.S meekens, 2005Sucrose-speci?c induction of anthocyanin biosynthesis in Arabidopsis requires the MYB75/PAP1gene.Plant Physiol.139: 1840–1852.
T uinstra,M.R.,G.E jeta and P.B.G oldsbrough,1997Het-erogeneous inbred family(HIF)analysis:a method for develop-ing near-isogenic lines that differ at quantitative trait loci.Theor.
Appl.Genet.95:1005–1011.
U ngerer,M.C.,S.S.H alldorsdottir,J.L.M odliszewski,T.F.C.
M ackay and M.D.P urugganan,2002Quantitative trait loci for in?orescence development in Arabidopsis thaliana.Genetics 160:1133–1151.
U ngerer,M.C.,S.S.H alldorsdottir,M.D.P urugganan and T.F.
C.M ackay,2003Genotype-environment interactions at quan-
titative trait loci affecting in?orescence development in Arabidop-sis thaliana.Genetics165:353–365.
V an O oijen,J.W.,1992Accuracy of mapping quantitative trait loci in autogamous species.Theor.Appl.Genet.84:803–811.
V an O oijen,J.W.,2004MapQTL5,Software for the Mapping of Quan-titative Trait Loci in Experimental Populations.Kyazma B.V.,Wage-ningen,The Netherlands.
V on K orff,M.,H.W ang,J.L eon and K.P illen,2004Development of candidate introgression lines using an exotic barley accession (Hordeum vulgare ssp.s pontaneum)as donor.Theor.Appl.Genet.
109:1736–1745.
X u,S.,2003Theoretical basis of the Beavis effect.Genetics165: 2259–2268.
904J.J.B.Keurentjes et al.
Y oon, D. B.,K.H.K ang,H.J.K im,H.G.J u,S.J.K won et al.,2006Mapping quantitative trait loci for yield com-ponents and morphological traits in an advanced backcross population between Oryza grandiglumis and the O.sativa ja-ponica cultivar Hwaseongbyeo.Theor.Appl.Genet.112:1052–1062.Z ou,F.,J.A.G elfond,D.C.A irey,L.L u,K.F.M anly et al., 2005Quantitative trait locus analysis using recombinant inbred intercrosses:theoretical and empirical considerations.Genetics 170:1299–1311.
Communicating editor:D.W eigel
Comparing a RIL and a New NIL Mapping Population905