Association Studies

1
Association Studies

• Gen 619

2
Bioinformatics

• From linkage to association
• Use assembled Human Genome websites to pick
  candidates in a linkage region
• positional candidates
• Find markers SNPs near or in genes of interest
• See if there's some Association

3
Association Studies
Principles and problems

• Case - control
• Simple
• Very Powerful
• Subject to spurious results
• Family-based
• Fairly simple
• Not quite so very powerful
• Can control for spurious association

4
Transmission Disequilibrium Test

• Spielman et al 1993
• McNemar 1947
• Heterozygous parent is p.5 for allele
  transmission
• Assumes no segregation distortion
• Check with unaffecteds

5
TDT for Quantitative Traits

• Allison (1997) TDTQ1-5
• TDTQ4 uses dichotomized phenotype and TDT McNemar
• TDTQ5 regresses of phenotype on offspring
  genotypes while controlling for parental mating
  types. AaAa AAAa Aaaa
• 2 Dummy variables for mating type
• 2 Dummy variables for offspring genotypes
• F-test 2df for offspring genotypes

6
Extensions

• Multiple Alleles
• Multiple Loci
• Logistic Regression
• Multiple Phenotypes (Manova)
• Imprinting (van den Oord 2000)

7
More extensions to QTDT

• Multiple offspring
• Large Pedigrees
• When parental genotypic information is missing

8
Quantitative Association Model
Each sib pair i has different MEANS
M
Geno1
Geno2
G1
G2
m
m
a
a
LDL1
LDL2
R
R
C
Geno1 Geno2 are OBSERVED genotypes scored -1 0
or 1 for aa Aa and AA respectively
9
Pair means differences
G Sib1
G Sib2
Pair Sum/2
Pair Diff/2
Mean Sib1
Mean Sib2
A1A1
A1A1
b
0
b
b
A1A1
A1A2
b/2
w/2
b/2w/2
b/2-w/2
A1A1
A2A2
0
w
w
-w
A1A2
A1A1
b/2
b/2-w/2
b/2w/2
-w/2
A1A2
A1A2
0
0
0
0
A1A2
A2A2
w/2
w/2-b/2
-b/2-w/2
-b/2
A2A2
A1A1
0
w
-w
-w
A2A2
A1A2
-b/2
-b/2w/2
-w/2
-b/2-w/2
A2A2
A2A2
-b
0
-b
-b
Within pair distance between homozygote w
between pairs it is b
10
Fulker Association Model
Multilevel model for the means
M
Geno1
Geno2
G1
G2
0.50
-0.50
0.50
0.50
S
D
m
m
w
b
B
W
1.00
-1.00
1.00
1.00
LDL1
LDL2
R
R
C
Cat's cradle
0.75
11
Applications of abaw model

• Neale M.C., S. S. Cherny, P. C. Sha3, J. B.
  Whitfield, A. C. Heath, A. J. Birley and N. G.
  Martin Distinguishing Population Stratification
  from Genuine Allelic Effects with Mx Association
  of ADH2 with Alcohol Consumption Behav Gent 1999
  29233-243  Fulltext (pdf)

12
Extensions of abaw model

• Posthuma D, de Geus EJ, Boomsma DI, Neale M.C.
  Combined linkage and association tests in mx.
  Behav Gent 2004 34179-196  Fulltext (pdf)
• Multi-allelic and dominance cases
• http//www.vipbg.vcu.edu/vipbg/neale-articles.shtm
  l

13
Potential for Joint Linkage and Association

• Allison DB, Neale M.C. Joint tests of linkage
  and association for quantitative traits. Theor
  Popul Biol. 2001 60239-251  Fulltext (pdf)
• Zhu G, Duffy DL, Eldridge A, Grace M, Mayne C,
  O'Gorman L, Aitken JF, Neale M.C., Hayward NK,
  Green AC, Martin NG A major quantitative-trait
  locus for mole density is linked to the familial
  melanoma gene CDKN2A a maximum-likelihood
  combined linkage and association analysis in
  twins and their sibs. Am J Hum Genet 1999
  65483-492  Fulltext (pdf)

14
Types of Association Study

• Alternative to Linkage
• Allele-frequency
• Case-Control
• Sullivan PF, Eaves LJ, Kendler KS, Neale M.C.
  Genetic case-control association studies in
  neuropsychiatry. Arch Gen Psychiatry 2001
  581015-1024  Fulltext (pdf)
• Family Based

15
Case Control Studies
Association

• Genotyping
• Inexpensive
• Reliable
• Alleles
• Functional
• Non-functional
• Markers
• May be alleles

16
Subject/Population
Case-control

• Unrelated
• Lifelong diagnosis
• Risk-period equivalent
• Subtype
• Control group (exposed)

17
Possible outcomes

• True negative
• False negative
• False positive
• True positive

18
True negative

• Satisfactory result
• Might get it for wrong reason

19
False negative

• Experimental bias
• genotyping error
• Chance
• Overmatching

20
False Positive

• Chance
• Incorrect prior
• Population stratification

21
True Positive

• Satisfactory result (very)
• Causal
• Non-causal
• Linkage disequilibrium

22
Summary
Association Studies

• Case-control is useful design
• May suffer from false positives/negatives
• Several common causes of errors
• Some ways to control
• Family-based association is good alternative

23
Conclusion

• Linkage still interesting
• Methodologically
• Substantively
• Association even more interesting
• Substantively
• Bioinformatics helps get from linkage to
  association
• Proliferation of statistical methods for
  association
• Dont forget about multiple testing!!!