Transcript definition of genetic disease

Allelic Association and
Transmission Disequilibrium Test
成大醫學院
病理科分子病理實驗室
分子診斷實驗室
呂政展 (x 2641)
Linkage Disequilibrium (allelic
association)
• Def: nonrandom association of
alleles at linked loci
• f(A, B)= f(A) x f(B)
•:
Haplotype
• Haplotype: a series of alleles found at linked loci on a
single chromosome. e.g,
– A33-Cw10-B58-DRB1*0301-DQB1*02(6.3%); A2-B46-Cw11DRB1*09-DQB1*0303 (orientals);
– A30-Cw5-B18(Bw6)-DRB1*0301-DQB1*02 (most frequent in
Mediterranean)
– A1-Cw7-B8(Bw6)-DRB1*0301- DQB1*02 (commonest in
European Caucasoids)
The Relationship of Disease Locus and
Marker Defined by Linkage Disequilibrium



The stronger the linkage disequilibrium, the closer
the marker is to the disease locus
Mapping and/ or gene identification using LD is
particularly useful in genetically unique or isolated
populations (Amish and Finnish populations)
LD rarely extends more than 1 cM from the
susceptibility locus, its detection indicates a
significant narrow down of candidate region
Allelic Association Studies

How it happens?



Direct biological action of the genetic polymorphism
Linkage disequilibrium with a adjacent susceptibility
gene
How useful in the analysis of genetic complex
disease?


Evaluation of candidate gene loci
Fine mapping of region that is indicated by linkage
analysis for follow-up studies
Association vs Linkage



Association and linkage: distinct from each other
Linkage: a specific genetic relationship between
loci
Association: Statistical statement about the cooccurrence of alleles or phenotypes
“Allele A is associated with disease D” :
if individuals who have D also have A more often
than would be predicted from the individual
frequencies of D and A in the population
Allelic Association


Def: A significantly increased or decreased
frequency of a marker allele with a disease
trait and represents deviations from the
random occurrence of the alleles regarding
the disease phenotypes
Linkage disequilibrium means allelic
association caused or maintained by tight
linkage
Linkage Disequilibrium in Genetic Analysis



Defining the ancestral haplotype of a disease
gene in relation to several marker loci
Fine-mapping of the disease gene even when
complete linkage ( = 0.0) is established in
the family studies
The slowness of linkage disequilibrium decay
make LD a useful mapping tool
Conditions for LD Mapping

Founder population



Geographically isolated population, traceable ancestry, most
individuals of current population are descendants of a few
individuals back  250 years ( 20 generations)
The prevalence of a genetic disease is derived from an allele
for a disease gene from a common ancestor
Both parents in many of the marriages within a founder
populations are heterozygous for a recessive disease gene,
will have a 25% chance of having an affected offspring
Rate of Decay of Linkage Disequilibrium:
time and distance-dependent





Dt = Do(1-)t
t : current generation number
Dt : current amount of disequilibrium
Do : disequilibrium at generation 0
: recombination fraction between loci
Decay in linkage disequilibrium
Significance of Allelic Association


Allelic associations reflect sharing of ancestral
chromosomes, only alleles at loci tightly linked to the
disease susceptibility locus will still be shared
For a locus showing recombination fraction (θ) with
the susceptibility locus, a proportion (θ ) of ancestral
chromosome will lose the association each generation,
and a proportion (1- θ) will retain it.


(1-0.01)44 = 0.64, loci 1cM apart
(1-0.03)44 = 0.26, loci 3 cM apart
Linkage Disequilibrium as a
Mapping Tool-I
Cystic fibrosis


mapped to 7q32, the candidate region was still
very large
XV2.c and KM19 marker within the candidate
gene show strong association between
(XV2.c*1, KM19*2) haplotype and CF
Linkage Disequilibrium as a
Mapping Tool-II
Nijmegen breakage syndrome (NBS)


Rare autosomal recessive disease characterized
by chromosomal breakage, growth retardation,
microcephaly, immunodeficiency and a
predisposition to cancer
Genetic linkage in families mapped NBS to an 8Mb regions between D8S271 and D8S270
Linkage Disequilibrium as a
Mapping Tool-II
Nijmegen breakage syndrome (NBS)



74 haplotypes related to a common ancestral
haplotype.
Some do not have this haplotype, possibly
carry independent NBS mutations.
Others share only part of haplotype, showing
the effect of recombination in distant
ancestors
Ancestral haplotype in
patients with Nijmegen
breakage syndrome
A: allele attributed to AH
51 unrelated patients
typed for microsatellites
Linkage Disequilibrium Mapping
I1
G1
E1
C1
A1
N
B1
D1
F1
H1
J1
K1
Mutation
N→M
I1
G1
E1
C1
A1
10
M
B1 generations
D1
F1
H1
J1
K1
I2
G3
E1
C1
A1
90
M
B1 generations
D1
F1
H1
J3
K4
I3
G2
E2
C3
A1
M
B1
D1
F2
H3
J4
K2


Jennings 1917 first developed the
concept of LD
Richard Lewontin (1964) first used LD
to measure D
Linkage Disequilibrium Coefficient
(D)(Lewontin 1964)

D = P11- p1q1 (if D significantly differs from zero, LD
is thought to exist)
locus A(1,2)



locus B(1,2)
P11 :observed frequency of the 1/1 haplotype
p1 : frequency of the allele “1” at the locus A
q1 : frequency of the allele “1” at the locus B


p1 = f(A1), p2 = 1- p1= f(A2)
q1 = f(B1), q2 = 1-q1 = f(B2)

Dmax = min(p1q2, p2q1)
Dmin = max(-p1q1, -p2q2)
D’ =D/Dmax

Alternatively,




D = h11- p1q1 , where h11 is the frequency of
the haplotype with the rarer allele at each
locus, and where p and q are frequencies of
the rarer alleles at loci 1 and 2, respectively
(Devlin and Risch, 1995).
D’ is positive when the rarer alleles at each locus are
associated and is negative when a common allele is
associated with a rare allele

Dmax = min[(pq, (1-p)(1-q)], for D <0

Dmax = min[(p(1-q), q(1-p)], for D >0
Determination of LD Coefficient
D =
d
n
bd cd
x
n
n
•a, b, c, d are the phenotype frequencies of the +/+, +/-, -/+
and -/- combinations of the alleles in each haplotype and
n is the sum of the a, b, c, d.
•HFij = Dij + GFi x GFj
•GF = 1 -
1  AF
Measure LD by D(2)
(Hill and Robertson 1968)

R or  =√2/N =
D/√(p1p2q1q2)

2 statistic= obtained
from 2 x 2 table


N: total number of
haplotypes in the
sample
p11
P12
p21
p22
Genome research 10: 1435-1444
Measure of LD-(3)
(Bengtsson and Thomson 1981)



= D/(q1p22)
q1: The population frequency of a
disease allele, B1
P22: the frequency of chromosome that
contain marker allele A2 and normal
allele, B2

 and D’ : give more reliable estimates
of physical distances than do D and R,
because D and R depends on allele
frequency
Quantitative Measurement of Linkage
Disequilibrium

Yule coefficient = (p1,1- p1,2)/(p1,1 + p1,2 － 2
p1,1 p1,2)(used in Huntington’s disease)
A1, A2
locus A


B1, B2
locus B
p1,1 = the frequency of allele A1 on the chromosome
p1,2
carrying allele B1
= the frequency of allele A1 on the chromosome
carrying allele B2
Allelic association around the locus for Huntington Disease
Acc I
Total Distance= 2500 kb
250 kb
Mbo I
Taq I
L19ps11
BS674
(n1.)(n.1)/N = expected values(600 x 300 /1000 =180)
Advantages of Association
Studies



Association studies are easier to conduct than
linkage analysis , no multicase families or special
family structures
LD is a short range phenomenon, if association
exists, it defines a small candidate region in
which to search for disease gene
Association is more powerful than linkage for
detecting weak susceptibility alleles
Possible Causes of Positive
Association-I



Direct causation: having allele A make you susceptible to
disease D; possession of A is neither necessary nor
sufficient to develop disease, but it increases the
likelihood
Natural selection: People who have disease D might be
more likely to survive and have children if they also have
allele A
Population stratification: population contains several
genetically distinct subsets. Both the disease D and allele
A happen to be particularly frequent in one subset (HLAA1 and chopsticks user in San Francisco bay area).
Possible Causes of Positive
Association-II


Statistical artefact: association studies often tests
a range of loci, each with several alleles, for
association with a disease. The raw p values need
correction for the numbers of genes tested
Linkage disequilibrium: If LD exist between the
disease D and the locus A, there should be a gene
near to the A locus that has mutations in people
with disease D.
Advantages and Pitfalls of
Association Study

Advantages



Not restricted to nonmendelian genetics
Easy to perform (case and control)
Pitfalls

Selection of controls is very crucial
( representative?!)
Probabilities calculated from Association
studies must be corrected for the number of
questions asked
Threshold of significance is p = 0.05/n (Bonferroni correction)
Relative Risk (RR)/Odds Ratio(OR)

Estimate how many times more the
carrier of a specific allele or haplotype
was likely to have a specific disease
associated with the particular genetic
marker studied
Calculation of RR





RR= a x d/ b x c
a: the number of patients positive for the allele or
haplotype
b: the number of patients negative for the allele or
haplotype
c: the number of controls positive for the allele or
haplotype
d: the number of controls negative for the allele or
haplotype
Calculation of RR (Relative Risk)
Risk factor
Present
Absent
Total
Cases
a
c
a+c
Controls
b
d
b+d
Sample
Total
a+b
c+d
n

Linkage disequilibrium mapping is
carried out after genetic linkage
between a polymorphic locus and
the disease gene is determined
The Genetic Distance() between a Marker
Locus and the Disease Gene that are in Linkage
Disequilibrium-I





Pexcess= Paffected-Pnormal/(1- Pnormal)= (1-  gq-1 )(1-  )g
= recombination fraction between marker and
disease loci
= mutation rate (1 x 10-6)
g= the number of generations after since the common
ancestor
q= world wide frequency of the disease ( =0.001)
The Genetic Distance() between a Marker
Locus and the Disease Gene that are in LD-II



Pnormal= the proportion of the marker allele in
normal chromosomes
Paffected = the proportion of the marker allele in
diseased chromosomes
Pexcess = measure of disequilibrium, the fraction of the
excess occurrence of a chromosome with the disease
gene and a marker allele compared with the
chromosome with the nondisease gene and the marker
allele
Functional SNPs in the lymphotoxin-α
gene that are associated with
susceptibility to myocardial infarction
(Nature Genetics 32: 650-654, 2002)



Linkage disequilibrium mapping
Haplotype analysis
D’= D/Dmax
65,671/92,788 = 70.8%
A cut-off P value of 0.01 for association in either
recessive and dominant mode

A SNP in intron 1 of LTA (252A->G) on
chromosome 6p21was associated with
myocardial infraction in the initial
screening




Construct a high-density SNP map for LD
mapping by sequencing 16 MI patients and
16 normal controls.
187 SNPs, 130 kb within 6p21
Select 120 SNPs (>10%) and genotyping 94
MI patients and 94 subjects from general
population
26 SNPs with minor allele frequency > 0.25
+ive
1
2
3
4


Located in one extended block of
intense LD, with D’ dropping off near
p5-1 and AIF1
They concluded that the gene
associated with susceptibility to
myocardial infraction was probably
located between these two loci
The Disatrophic Dysplasia Gene Encodes
a Novel Suflate Transporter: Positional
Cloning by Fine-Structure Linkage
Disequilibrium Mapping
Cell 78: 1073-1087, 1994
Diastrophic dysplasia (DTD)
(彎曲發育不全)




One form of Osteochondrodysplasia
Constitutional disorders of skeletal
system result in disturbed growth
and/or density of bone
First described by Lamy and Maroteaux
in 1960
Autosomal recessive
Diastrophic dysplasia)(DD)
(彎曲發育不全)



Short-limbed short stature, kyphoscoliosis（脊柱
後側彎), generalized dysplasia of the joints,
peculiar flexion limitation of the finger joints,
hitchhiker thumbs, metatarsus adductus (蹠內收)
deformity of the feet, and deformation of the ear
lobes and cleft palate
Joint changes are progressive in nature
Osteoar’throses(非炎性骨關節病) and
contractures develop at an early age
Diastrophic dysplasia)(DD)
(彎曲發育不全)





Patients are severely physically handicapped and
need repeated corrective orthopedic surgery
Increased mortality in infancy due to respiratory
difficulties and spine anomalies
Intelligency and sexual development are normal
Life expectancy is not clearly shortened
Abnormal cartilage, number and morphology of
chondrocytets, the organization of collagen fibrils,
and the deficiency of sulfated proteoglycans
Diastrophic dysplasia and D5S72 is
Genetically Closely Linked
No recombination between DD and D5S72 (Z()= 7.37)
Principle of Linkage Disequilibrium (LD)
Mapping

Among a collection of chromosomes
carrying the same ancestral mutation,
genetic markers nearest the disease
gene will have recombined least often
and thus should show the highest degree
of allelic association with the disease on
such chromosomes
How to Do Linkage Disequilibrium
Mapping



Determine the haplotypes of disease-bearing
chromosomes for an extremely dense
collection of genetic markers
Use these haplotypes to identify a subset of
chromosomes likely to carry a common
ancestral mutation
Find the genetic markers that show the
strongest allelic association with the disease
on these chromosomes
Genetic Linkage studies of DTD



18 multiplex and 59 singleton families (a total
of 144 DTD-baring chromosomes)
DTD was initially localized to the interval
between GRL and SPARC on chromosome 5q
CSF1R gene: 1-1 haplotype at these two marker
loci was present on 95% of DTD-baring
chromosomes, but only 4% of normal
chromosomes (RFLP analysis)
Physical map shows locations and
direction of transcription




AC3: STS from the 3’ end of CSF1R
BT1: CA-repeat (SSR)
STS content mapping of the clones
Known genomic organization of
PDGFRB-CSF1R
Genetic map ordered by
recombinational mapping



DTD was estimated to lie about 0.064 cM from
CSF1R
Genetic map order of each pair of adjacent loci
supported by obligate crossovers with exchange of
flanking markers either in CEPH pedigrees or DTD
families
GRL-ADRB2-D5S413-D5S372-BT1-CSF1RPDGRRB-RPS14-SPARC-D5S72
Physical map shows locations and
direction of transcription




AC3: STS from the 3’ end of CSF1R
BT1: CA-repeat (SSR)
STS content mapping of the clones
Known genomic organization of
PDGFRB-CSF1R
Number of Normal and DTD chromosomes carrying
Alleles from Finnish Ancestral DTD Haplotype


Most DTD-bearing chromosomes carrying a
single predominant allele at each of the 11
markers studied
95% of
DTD-bearing chromosomes with haplotype
“1-1-1-1-1-1-1-1-1-1-1”
Linkage Disequilibrium Mapping





P
= (P affected – P normal)/(1-P normal)
≒α(1-θ)g
P directly related to proximity to the disease
locus
α = proportion of disease-bearing chromosomes
g = the number of generations since the
introduction of ancestral chromosome
excess
Linkage Disequilibrium in DTD Region
Clues Regarding the Location of
DTD: Proximal to CSF1R



P excess falls rapidly a the PDGFRb-CSF1R, but
remains high at the marker BT1 proximal to
CSF1R.
Haplotype pattern of 144 DTD chromosomes
Single DTD chromosome lack ancestral allele
at the genetic markers within the PDGFRB
and CSF1R and present at the more proximal
genetic markers BT1 and D5S372.
Gene Identification of Region 100
Kb Proximal to CSF1R




Exon trapping of P1-1013, P1-1014
Direct cDNA selection
Analysis of genomic fragments from this region
Clones (JH10140B, 0.8 kb fragment from
centromeric end of P1-1014, show strong aa
similarity to the 5’ end of rat gene (sat-1, sulfate
transporter) from BLAST program
Expression Profile of Candidate DTD Gene
Northern blot
analysis
Probe: 0.8 Kb JH10140B
DTD 8.4 kb mRNA is Missing from DD Patients
DTDST gene contains two exons
separated by 1.8 kb intron
DRA gene (down-regulated adenoma) from subtractive hybridization between
colon and colon carcinoma
DTDST Shows 12 Transmembrane
Helices (hydropathy analysis)
Two potential N-glycosylation sites at the N-terminal extracellular loop
Confirmation of Mutation
Normal
Normal
SP Family
FV Family JD IM
Codon 575 (AAG→AG, Dde I eliminated)
SA(AG→AC, BsaAI created)
100 bp(Dde I+) → 40 bp + 60 bp
1000 bp→ 2 x 500 bp
Summary of Mutations
Transmission disequilibrium test (TDT)




Association studies with internal controls
Couples who have one or more affected
offspring, it is irrelevant wheterh either
parents is affected or not
To test whether marker alleles M1 is
associated with the disease
Those parents who are heterozygous for M1


Let a be the number of times a
heterozygous parent transmits M1 to
the affected offspring, and b be the
number of times the other allele is
transmitted.
TDT test statistic is (a-b)2/(a+b)
Stevens-Johnson Syndrome (SJS)
or Toxic Epidermal Necrosis (TEN)



In 1922, Stevens and Johnson described two
children with fever and stomatitis, severe
disseminated conjunctivitis and cutaneous
eruptions
The reaction develops within 1-4weeks from the
beginning of drug therapy
Mucosal lesion includes lesions of the mouth,
eyes, GI, respiratory tract, anus and vagina
Drugs associated with SJS and
TEN






Sulfonamides,
anticonvulsant agents
Allopurinol, are the most consistently
associated with SJS and TEN
Nonsteroidal antiinflammatory drugs
(NSAIDs)
Analgestic agents
Nonsulfonamides antibiotics,
controversial?
Typical pattern of StevensJohnson syndrome (SJS)
Blisters develop on widespread purpuric macules
N Engl J Med 1333: 1600-1607, 1995
Typical pattern of toxic epidermal
necrosis (TEN)


Severe forms of SJS
Extensive skin
detachment and a
poor prognosis
(death rate 30~40%)
Blisters and wrinkled areas result from full-thickness necrosis o
A marker for Stevens-Johnson Syndrome







carbamazepine, treatment for seizures(癲癎)
44 CBZ-SJS patients
101 CBZ-tolerant (93 Normal controls)
Diagnostic criterion of SJS/TEN were based
on clinical morphology
SJS: skin detachment of 10%,
TEN: skin detachment of 30%,
Overlapping SJS/TEN: skin detachment of 1030%
Nature 428: 486, 2004
Genotyping methods



HLA-A, -B, -C, DRB1(sequence-specific
oligonucleotide probe, SSOP)
cytochrome p450 (157 Single nucleotide
polymorphism);
What is single nucleotide polymorphism ?
ATG(M)→ATA(I); TTC(F) →ATC(I))
Nature 428: 486, 2004
Nature 428: 486, 2004