Methods in Genetic Epidemiology

Lindon Eaves

VIPBG

March 2006

Link to basic genetics

http://psych-www.colorado.edu/hgss/hgssfigures/hgssfigures.htm#Chapter10

“Epidemiology”

Identification of “causes from clusters”

Clusters and Causes

• Sources of water (cholera)

• Insect vectors (malaria)

• Working in mines (silicosis)

• Toxic waste (some cancers)

• Hypertension (ethnicity)

• Etc.

THE “FAMILY”

A primary cluster

                                                                                                                                                                                                   

Large pedigree from the isolate with a founder couple born in approximately 1650. Circles represent females; squares represent males. Blackened circles and squares represent individuals with at least two children affected with schizophrenia.Am J Hum Genet. October 1999; 65(4): 1114–1124. Published online September 3, 1999. Copyright © 1999 by The American Society of Human Genetics. All rights reserved.

Insert Kuru pedigree here!

Draw your own pedigree

http://www.genesoc.com/nutrition/pedigree.htm

Familial disorders

• Familial hypercholesterolemia• Bovine spongiform encephalopathy• Schizohphrenia• Kuru• Colon cancer• Mental retardation• Breast cancer• Etc. etc. etc.

Family clustering may be genetic or environmental

“Genetic Epidemiology”

Systematic use of familial clustering to resolve and identify genetic and environmental causes of disease

Questions in GE

• Is “it” familial?• Is it “genetic” (G), “environmental” (E) or both

(G+E)?• What are relative contributions of G and E?• What kinds of genetic effects are there (one,

few, many, additive/epistatic)?• Effects of population structure, ethnicity,

stratification, mate selection?• What is role of E? Which? GxE interaction?• Which specific genes are involved?

P

G4G1 G2 G3 E1 E4E2 E3

GE

P1 P4P2 P3

G’4

E’4E’1G’2 E’3

G’1

Measured Genotypes Measured Environments

Outcome Phenotype

Endophenotypes

TIME?

P5

G’5

E’

= Pathway blocked by mutant gene

A

B

BC

A

aa

bb

aa bb

Sequential (“complementary”) genesParallel (“duplicate”) genes

Answering the questions:some approaches

Is it familial?

• Risk to relatives (schizophrenia, colorectal cancer, mental retardation, hypertension)

• Correlation between relatives (IQ, diastolic blood pressure)

• Same trait may be category or dimension (hypertension = DBP>90mmHg)

Causes of Family Resemblance

Path diagram for the effects of genes and environment on phenotype

P

G E

Measured variable

Latent variables

Genotype Environment

Phenotype

h e

r

M

C

F

Em Gm Gf Ef

Ec Gc

Mother Father

Offspring

h e

e h h e

r r

f f

½ ½

m

Family Resemblance

-14 -10 -6 -2 2 6 10 14

IHT

0.00

0.04

0.08

0.12

0.16

Distribution of stature corrected for age and sex(Inches from mean)

Probability ofOutcome

0

1

0 t

0.5

Liability

-

Polygenic Inheritance (Fisher, 1918)

0 1 2 3 4 5

Y1

0.0

0.1

0.2

0.3

0.4

Distribution of scores produced by two genes(N=1000 subjects)

-2.5 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 5.5 6.5

S1

0.0

0.1

0.2

0.3

0.4

The "smoothing" effect of the environment(N=1000 subjects, 2 gene model)

75 79 83 87 91 95 99 103 107 111 115 119 123

Y1

0.00

0.02

0.04

0.06

Continuous distribution of polygenic trait (100 genes with small cumulative effects)

b.

c.

a.

Estimating G and E (1)

Twin Studies (Galton 1865)

Figure 9. The correlation between twins for stature (data from the Virginia Twin Study of Adolescent Behavioral Development).

-10 -5 0 5 10

HTDEV1

-12

-7

-2

3

8

13

HT

DE

V2

Scatterplot for corrected MZ stature

r=0.924

-16 -11 -6 -1 4 9 14

HTDEV1

-20

-10

0

10

20

HT

DE

V2

Scatterplot for age and sex corrected stature in DZ twins

r=0.535

Stature in Twins

Path diagram for twin resemblance

P1 P2

E1

A2A1

C1 C2

E2

1

g

c c

e e

a a

Twin 1 Twin 2

Phenotypes

Four scenarios

00.10.20.30.40.50.60.70.80.9

No G No C G and C G and I

MZDZ

Twin Correlation

Causes of Variation

Estimating G and E (2)

Adoption studies

GAC EAC GBC EBC

GAF EAF GAM EAFEBFGBF GBM EBM

BF

BCAC

BM

AF

AM

p p

p

p1/2

1/2f

m

Biological Parents Adopting parents

Adopted Child Biological Child of Adopting Parents

Other approaches

• Separated Twins (e.g. Shields, 1966)

• Twins and parents

• Children of Twins

• Extended twin kinships

• Combinations of Methods

Identifying environments

Measure environment in family, twin or adoption study

(e.g. twins discordant for exposure, characteristics of

foster parents etc.)

Finding the genes

• Linkage studies – extended pedigrees, sib pairs

• Association studies (case-control, family-based – e.g. TDT, sib-pair)

“Linkage”

                                                                                                                         

Genes that start together stay together – the closer they are, the more they stay together

Linkage

Are relatives who are more alike for a “marker” more alike in their

phenotypes?

Q1 Q2

G1 G2

E2E1

T1 T2

½

h h

q q

e e

QTLGENES

ENVIRONMENT

PHENOTYPE

Linkage: Sibling Resemblance

Sib-pair similarity and linkage

0 10.5

SiblingCorrelation

0

Q

R

r

r=R+Q

Linkage – plusses and minuses

• PLUSES• Depends only on

marker location, not effect of marker on phenotype

• Doesn’t require many markers to cover genome (100s)

• Hard to “invent” linkage - robust

• MINUSES• Works best for simple

(few-gene) traits• Gene effects need to

be big• Specific localization

poor (i.e. many genes under linkage “peak”)

Association

Do different forms of gene (“alleles”) have different

phenotypes?

Association: Pluses and Minuses

• PLUSES• Statistical power• “Tight” localization• Can use “candidate

genes”

• MINUSES• Association may not be

causal (e.g. “linkage disequilibrium”, population stratification) – but can control/eliminate

• Needs large number of markers for genome-wide study (?500,000+)

P

M3

M1

M2

M4

Genome

Genome

Phenotype

Marker-Phenotype Association

Controlling for stratification

• Analyze within strata (ethnicity, SES etc)

• Use random genes to test, characterize and eliminate

• Family-based association – TDT, sib-pairs etc.

Complications – can’t always do genetics without environment

• GxE Interaction

• G-E correlation

• G x Age interaction

GxE Interaction

• Genes control sensitivity to the environment (some environments only affect particular genotypes)

• Environments modulate expression of genes (some genes only expressed in particular environments)

b. “Scalar” GxE

Environment (E)

PhenotypeGenotypes

c. “Non-scalar” GxE

Environment (E)

PhenotypeGenotypes

a. No GxE

Environment (E)

PhenotypeGenotypes

Analyzing GxE

• Family resemblance depends on environmental exposure

• Effect of gene contingent on environment (or vice-versa)

Genetic Variance and Shared Life Events in Adolescent Females (Silberg et al.,

1999)

00.5

11.5

22.5

33.5

44.5

0 1 2+

Number of life events

Gen

etic

Var

ian

ce

DepressionAnxiety

0

5

10

15

20

25

30

35

40

No Probable Definite

Low MAOA

High MAOA

Figure: Prevalence of Conduct Disorder as a function of MAOA activity and exposure to environmental adversity

(Foley et al, 2004)

G-E Correlation: Environmental exposure

caused/influenced by genes

• “Active/Evocative” - environment depends directly on genes of individual (e.g. own smoking”)

• “Passive” – environment depends on genes of relatives (e.g. parental smoking)

Analyzing rGE

• Include environmental measures in twin, adoption and family studies – build and test path models.

G x Age Interaction

• Genetic control of age of onset

• Different genes expressed at different ages

• Rates of growth/change depend on genes

- A bit like GxE in some ways.