QUANTITATIVE TRAIT LOCI (QTL) MAPPING

Gurbachan S. Miglani
Visiting Professor

SCHOOL OF AGRICULTURAL BIOTECHNOLOGY

PUNJAB AGRICULTURAL UNIVERSITY
LUDHIANA- 141004
 QUALITATIVE vs. QUANTITATIVE TRAITS

QUALITATIVE TRAITS QUANTITATIVE TRAITS

No. of genes involved A few Many
Phenotypic differences among individuals Discrete Continuous
No. of phenotypic classes A few Range of phenotypes
Effect of environment on phenotype No effect Lot of effect
Effect of individual gene on phenotype Large Small
Observation of phenotypes Visual Statistical parameters
 Examples of Qualitative
and Quantitative Traits

QUALITATIVE TRAITS QUANTITATIVE TRAITS

Round and wrinkled seeds of pea Kernel Colour in Wheat

Horned and hornless condition in cattles Weight of grains

Black and white coat of guinea pigs Eye colour in Man

Rolling tongue in Man Height of plants

Humans' ABO blood groups Skin Colour in Man

 Early Studies on Quantitative Traits
DISCOVEROR YEAR STUDIED
W. Johannsen 1903 Seed weight in beans – Pure line theory
G.V. Yule 1906 Polygenic inheritance
H. Nilsson-Ehle 1909 Kernel color in wheat
E.M. East 1910 Corolla length in Nicotiana longiflora
R.A. Emerson and E.M. East 1913 Ear length in corn
K. Sax and K. Ramusson 1923 Seed color in Phaseolus vulgaris – Linkage between
qualitative and quantitative characters
K. Mather 1943 Term polygene

Quantitative traits are controlled by a action and interaction of a large number

of genes each having an additive effect on the phenotype of an individuals

Such traits are smoothened by environment

BREAKTHROUGH DISCOVERIES

DNA Markers
UNDERSTANDING
GENETIC BASIS OF
QUANTITATIVE
TRAITS

Statistical Methods
& Tools
QUANTITATIVE TRAIT LOCUS (QTL)

A chromosomal region (stretches of DNA)

linked to or associated with a marker gene
which affects a quantitative traits

Mapping of QTLs
Determining the position of a locus causing
quantitative variation in the genome
 Objectives of QTL Mapping

 To identify the regions of the genome that affect the trait of interest
 To analyze the effect of the QTL on the trait
 To provide understanding of individual gene actions and interactions
 To enable positional cloning of the gene

 To assist in knowledge based breeding

 Principle of QTL Mapping
Detection of QTLs is based upon the association of polymorphic
markers with polymorphic genes controlling the trait of interest

1. M1 and M2 are molecular 2. M1 marker and the QTL are 3. No marker is polymorphic,

marker loci that flank a QTL  both polymorphic the QTL is polymorphic

Location of QTL
1. Hypothetical
2. Can be detected
3. Not detectable
4. Can be detected
5. Not detectable
Miglani, G.S. 2016. Genetic
Engineering Principles, Procedures
4. M3 is polymorphic, the 5. M1 and M2 are polymorphic, and Consequences. Narosa
QTL is polymorphic QTL is not polymorphic Publishing House, New Delhi .
 Components of QTL Mapping

1 Development of mapping population

2 Phenotypic evaluation 3 Marker evaluation

4 QTL statistical analysis

Analyze combined phenotypic
and marker data

Identify and characterize QTLs

 Mapping Populations

 Biparental populations

Backcross P1 P2
population

BC BC
F1

DH

F2 population
F2 Doubled haploid
5-6 (DH) population

Recombinant Inbred
lines (RILs)

 Multiparental populations

 Natural populations
 Comparing Different Mapping Populations
F2/F3 RILs Back cross DH
• Quick to create • Lots of • Amenable with • Quick to create
• Both additive and recombination trait introgression • Immortalized and
dominance effects • Immortalized and breeding easily replicated
PROs

easily replicated • Moderate and shared

and shared recombination
• Not immortalized • Takes years to • Less informative • Limited
• Limited recombination create recombination
• Limited ability to • Not amenable for • Costly and
replicate all species or laborious
crosses • Additive effects
CONs

• Additive effect evaluation only

evaluation only

Immortalized = famous; remember for a very long time; stable

Size of the Mapping Population

Depends on:

o Type of mapping population

o Genetic nature of the target trait

o Objectives of the experiment

o Resources available
 Evaluation
Multiple Measurement 
 methods Replicated 
locations or of multiple  Randomized
 and growing  designs
 years plants per line
conditions; uniform
Recording phenotypic data
2. Phenotypic Evaluation
 calculation
• End product; set of mean values for each line for ea
Mean 
ch trait
analysis
• Two traits correlated; same QTLs influences
• Linked QTLs
Correlation 
estimates
• Greater the proportion of total variability
due to genetic causes
Heritability
(ANOVA)
• Significant differences among genotypes Analysis of variance
distribution
• Normal distribution Frequency  Phenotypic data analysis
 3. Molecular Marker Evaluation

DIFFERENT TYPES OF MARKERS

Hybridization-based Polymerase chain Sequence-based

Reaction-based
Restriction fragment Single-nucleotide
Length polymorphism Sequence-specific repeats polymorphism
(RFLP) (SSRs) (SNP)

 To obtain complete genome coverage
 More densely spaced markers
 Parental Screening

Polymorphic markers; 
useful  in mapping the
population

Polymorphisms:
Size differences of DNA bands
Presence vs. absence of a DNA band
Evaluating the Population For Markers

Markers

F1
P1 P2

Mapping
population

MARKER DATA

Different genotypic groups

 Linkage Mapping

Screen the mapping population using polymorphic

molecular markers (genotyping)

Analyze the segregation patterns

for each of the markers

Genotypic data is then analyzed

using MAPMAKER , etc

Construction of a linkage map

Banding Patterns for Various Populations

Expected segregation ratios for markers in different population types

Collard BCY et al, 2005. Euphytica 142: 169-196

Collard BCY et al, 2005. Euphytica 142: 169-196
4. QTL Statistical Analysis
 A. Single-marker Analysis (SMA)/Single-point Analysis (SPA)

 Divide the population based on the segregation of individual marker loci

 Average the phenotypic trait of these classes
 Compare the trait means of different genotypic classes for each marker individually
 Determine statistically significant differences by ANOVA, Regression
• Each marker-trait association test; independent of information from all other markers
• Chromosome with n markers offers n separate single-marker tests

A marker Putative QTL

 “Is the mean trait value for all plants with the Parent A marker
pattern significantly different than the mean value of plants with the Parent B pattern?”

QTL linked to the trait of interest QTL not linked to the trait of interest

Collard BCY et al, 2005. Euphytica 142: 169-196

 Results Obtained From Single Factor Anova

1. A rough estimate of • A QTL is inferred to be located close to

the QTL position the most significant marker

Location on Parent contributing

Marker chromosome P-value R2 (%) higher value allele
A 18 cM <0.0001 6.3 C-306
B 19 cM <0.0001 6.7 C-306
C 23 cM 0.0018 6.2 C-306
D 28 cM <0.0001 10.8 C-306
E 37 cM 0.0005 5.3 C-306
 2. Measure of
• Absence of a marker-trait association
statistical • The lower the P-value higher the probability
significance:  that a QTL exists
P-value

• Percent of the total phenotypic variance for

3. % R2 the trait that is accounted for by a marker

• Mean values for the marker classes compared;

4. Source of the most favorable mean- source of the desired QTL
favorable allele  allele
 Mean of A marker class – Mean of B marker class
5. Estimates of   2
additive and
dominance (Mean of A class + Mean of B class)
effects Mean of heterozygous (H) class – 
  2

Limitations of Single Factor ANOVA

 Does not determine; direction of the QTL from the marker

 No information; marker association; how many QTLs
 B. Simple Interval Mapping

 QTL presence every 2 cM between each pair of adjacent markers
 Position of a QTL and the size of its effects are estimated more accurately
 At each test position, the SIM method calculates a LOD score
 Chromosome with n markers offers n-1 separate tests of marker-trait associations

r
r1 r2
A B
marker marker
Putative QTL
 QTL Mapping Statistics

LOD Score

LOD (logarithm of the odds ratio) =

Probability of observing the data under linkage

Probability of observing the same data under no linkage

 Key value = 3; this means 1000 more likely under linkage

 LOD scores are plotted along the chromosome map

 The threshold depends upon marker density of linkage map

 Most likely QTL position interpreted; point where the peak 
LOD score occurs

http://www.maswheat.ucdavis.edu
Results Obtained from Simple Interval Mapping

1. Estimate of QTL position
2. Measure of statistical significance
3. Percent variance explained (%R2)
4. Source of desirable alleles
5. Estimates of additive and dominance effects

Limitations of Simple Interval Mapping

1. Indicated positions of QTLs are sometimes ambiguous, 
or influenced by other QTLs
2. Difficult to separate effects of linked QTLs
 C. Composite Interval Mapping

• Combines interval mapping with multiple regression

•  Reduces background “noise” that can effect QTL detection

• More precise and effective

Results obtained from Composite Interval Mapping 

1. Estimate of QTL position
2. Reduce background noise
3. Measure of statistical significance
4. Percent variance explained (% R2)
5. Source of desirable alleles
6. Estimates of additive and dominance effects
 Limitations of QTL Analysis

• Information on QTL locations and effects; cannot 
be transferred to another population

• QTL analysis; expensive in time and materials

• Detection of QTLs with minor Effect

• QTLs in repulsion phase linkage

 Progress in QTL Mapping
Crop QTL Trait Reference
Tomato fw2.2 Fruit size Alpert et al. (1995)
Ovate Fruit shape Ku et al 1(999)
Wheat GPC Grain protein Joppa et al (1997)
Distelfied et al (2004)
Cbf3 Frost tolerance Vagujfalvi et al (2003)
QYrdr.wgp-5BL.1 Stripe rust Hou et al (2015)
resistance
QYrdr.wgp5DL
QYrdr.wgp-6BL
Brassica COL1 Flowering time Osterberg et al (2002)
E1 Euric acid content Peleman et al (2005)
Rice Gn1a (Grain number 1a) Grain number Ashikari et al (2005)
DEP1 (Dense and errect panicle 1) Grain number Huan et al (2009)
WFP (Wealthy farmers’ panicle) Grain number Miura et al (2010)
IPA (Ideal plant architecture) Grain number Jiao et al (2010)
GS3 Grain weight Fan et al (2006)
GW2 Grain weight Song et al (2007)
GW5 Grain weight Shomura et al (2008)
Weng et al (2008)
Crop QTL Trait Reference
Rice Hd3a Heading date Kojima et al (2002)
Izawa et al (2002)
Tamaki et al (2007 )
Ehd1 Heading date Doi et al (2004)
Ghd7 Grain number, plant height and Xue et al (2008)
heading date
sd1 Plant height Sasaki et al (2002)
SCM2 Lodging resistance Ookawa et al (2010)
SKC1 Salt tolerance Ren et al (2005)
Sub1a Submergence tolerance Xu et al (2006)
sh4, qSH1 Seed shuttering Li et al (2006)
Konishi et al (2006)
GIF1 For increased Grain filling Wang et al (2008)
Hd1 Heading date Yano et al (2000)
Hd6 Heading date Takahashi et al (2001)
Maize %LCA Root traits Burton et al (2015)
AA Root traits
RXA Root traits
qKW10a Zhang et al (2014)
qKW7a Kernel weight
qKW5
 A CASE STUDY IN WHEAT

PLoS ONE 2015,10(5): e0126794. doi:10.1371/journal.pone.0126794

Cultivar Trait Population used Strategy Markers

 Winter  Resistance to  F8 RIL derived from  CIM  SSRs
wheat stripe rust
Druchamp Druchamp  SNPs
(resistant)
 Michigan x
Amber Michigan Amber
(susceptible)
94 RIL populations

Phenotyping for stripe rust: multiple years

 Under natural infection
 Selected Pst races under controlled greenhouse

768 (240 polymorphic) SSR markers

9000 (2535 polymorphic) SNP markers

Linkage map construction: 156 SSR and 2535 SNP

  Linkage maps and position of three all-
stage resistance QTL
 Markers with prefix X: SSR markers

IWA: SNPs markers

Closely Race Variation
QTL linked explained
marker (%)
QYrdr.wgp-5BL.1 IWA6271 PST-29 36.04
PST-45 30.57
PST-100 5.47
QYrdr.wgp5DL IWA8331 PST-35 11.94
PST-45 9.27
QYrdr.wgp-6BL.1 IWA3297 PST-35 13.07
PST-100 20.36
PST-114 18.66

Hou, L et al. 2015. Plos one 10(5): e0126794

Hou, L et al. 2015. Plos one 10(5): e0126794
  32 linkage groups
 Assigned to 18 wheat chromosomes using
reference SSR and SNP linkage maps
(GrainGenes database http://wheat.pw.
usda.gov)
 11 QTLs mapped
Results o 3 race-specific all-stage resistance
QTL QYrdr.wgp-5BL,QYrdr.wgp-
5D and QYrdr.wgp-6BL.1 mapped
o 8 QTLs for high temp adult plant (HTAP)
resistance
• Two on chromosomes 1BL and one each
on 1DS, 2BL, 3AL, 5AL, 5BL and 6BL
 Applications of QTL Mapping
Sequence of the genes
underlying QTL and functional
characterization

1
QTL
Cloning
2 5
Estimated number of QTL, To combine two or
Magnitude of estimated
Genetic
Control of QTL more QTLs into the
additive, dominance, and Pyramiding same line –
Epistatic effects Trait “stacking”
QTL
Mapping

3 4
Marker-
AB-QTL
Assisted
Analysis
Use of markers Selection Simultaneous
associated with identification and
desired QTL transfer of QTL
References

Miglani, G.S. 2007. Advanced Genetics. Second Edition. Chapter 22.

Copublished by Alpha Science International, Oxford, U.K. and Narosa
Publishing House, New Delhi, India.
Miglani, G.S. 2008. Fundamentals of Genetics. Chapter 21. Copublished by
Alpha Science International, Oxford, U.K. and Narosa Publishing House, New
Delhi, India.
Miglani, G.S. 2015. Essentials of Molecular Genetics. Chapter 27. Copublished
by Alpha Science International, Oxford, U.K. and Narosa Publishing House,
New Delhi, India.
Miglani, G.S. 2016. Genetic Engineering Principles, Procedures and
Consequences. Chapters 7 and 21. Copublished by Alpha Science
International, Oxford, U.K. and Narosa Publishing House, New Delhi, India.