Behavioral genetics 2

Multifactorial/polygenic inheritance

Multiple genes at different loci influence phenotypes (small and equal with additive effects)

How is multifactorial/polygenic inheritance different from Mendelian inheritance?

Mendelian inheritance describes a trait that is determined by only one gene.

Quantitative phenotype

genetic (multifactorial: multiple genes) and environmental contributions

What is the most important difference between MZ and DZ twins to a behavioral geneticist?

MZ twins share 100% of their genetics

DZ twins share 50% of their genetics (siblings)

How do MZ twins occur?

a single, fertilized egg (zygote) divides into two embryos, due to unknown causes

How do DZ twins occur?

when there are multiple ovulations (2 eggs and 2 sperm) due to hormonal treatment, maternal age effect, mother’s ancestry, assisted reproductive technologies

Monochorionic vs dichorionic

MZ twins can have their own chorion or share one. DZ twins are always dichorionic. A chorion forms the placenta. 

Heritability

the proportion of phenotypic variance accounted for by genetic differences

biometric decomposition

phenotypic variance = genetic component + environmental component

1 = (a2 + d2 + i2) + (c2 + e2)

a2

additive genetic variance (effect of single alleles aggregated over multiple loci) 

d2

dominance (effect of combined alleles inherited over multiple loci)

c2

common/shared environment

e2

non shared environment

broad sense heritability

proportion of phenotypic variance attributable to all genetic factors

narrow sense heritability

proportion of phenotypic variance attributable to additive genetic factors only

What is heritability useful for?

-predicting response to selection

-indexing the extent to which genetic factors currently predict phenotypic differences

-determining evidence for gene mapping effects

-indicating whether genetic research strategies are likely to be successful for a given trait

What is heritability NOT useful for?

-determining if a trait is genetically fixed (NOT a fixed biological constant)

-explaining differences in populations or across time periods

-interpreting heritability as applying to an individual (it’s a population statistic, not an individual measure)

examples of heritability

Ex. Heritability of height: ~80% of variance in height is attributable to genetic factors, but the environments can still change height

Ex. Heritability of reading: Sweden does not have a reading curriculum until first grade, so the environment is more important (reading ability is less in Sweden because of the environment)

What is the Falconer model?

The additive genetic components are most important

1 = a2 + c2 + e2

variance due to genetics

a2 = 2(r_MZ -r_DZ) 

variance due to shared environment

c2 = 2r_DZ - r_MZ

variance due to non shared environment 

e2 = 1- r_MZ 

How do relatives share biometric components?

MZ twins are more genetically similar to DZ twins due to additive genetic effects. There is little evidence for non additive genetic effects. MZ twins are not identical, due to variance from the non shared environment. The shared environment is equally similar for MZ and DZ twins. Both siblings and parent-offspring share ~50% of

What is the classical twin study?

Classic twin studies are done to obtain converging evidence for the existence of genetic and environmental influences, MZ and DZ twins reared together are compared. If genetics is the major source of individual differences, MZ twins will be more similar. If rearing environment is the major source of individual differences, MZ twins and DZ twins will be equally similar

Assumptions of the classic twin study design

Assumption representativeness

Equal environments assumption

Assumption representativeness

twins are psychologically like the non twin population (generalizable)

there is almost no difference between twins and non twins in personality measures or cognitive ability

equal environments assumption

the environmental similarity of MZ twins is not greater than the environment similarity of DZ twins

What are the basic findings of the Polderman et al. meta analysis of twin studies? 

1. Twin similarity for psychological traits is not that different from their similarity for physical traits

2. MZ twins are more similar than DZ twins, there is little evidence for non-additive genetic effects

3. MZ twins are not perfectly similar (due to non shared environment

4. Estimates of shared environmental effects tend to be small for most traits

   1. mean heritability across all traits was ~49% indicating that about half of the variation in human characteristics can be attributed to genetic differences

Adoption study merit

Adoption studies remove the confound of nuclear families (parent-offspring resemblance can arise for both genetic and environmental reasons) because the genetic and environmental contributions to parent-offspring resemblance are separate in adoption studies.

Selective placement

matching adoptive parents with adoptive children based on similar characteristics (child’s environment is correlated with biological predispositions) or creating an advantaged environment, which can bias biological and environmental comparisons

MISTRA 

Minnesota Study of Twins Reared Apart (removing environmental similarity) provided convergent support for the three laws of behavioral genetics

3 Laws of Behavioral Genetics

1. all human behavioral traits are heritable (a2)

2. the effect of being raised in the same family is smaller than the effect of genes (c2)

   1. the non shared environment is the largest source of environmental variance (e2)

Quasi intervention in adoption studies

Adoption studies naturally separate genetic and environmental influences without requiring experimental manipulation. Adoption is a “natural experiment” that mimics random assignment.

Adoptive homes are typically advantaged and studying adopted individuals can provide an estimate of the impact of advantaged rearing

Influence of shared rearing environment

Minimal impact on personality/cognitive abilities, but does influence VALUES such as social attitudes, religiousness, and political beliefs

Gene environment interaction

the magnitude of the genetic effect depends on the environment and the magnitude of environmental effect depends on the genetic background (intersecting lines on a graph); genes and environments combine to influence a trait (responsiveness)

examples of gene environment interactions

Depression: how many stressful life events does it take for genotype (risk alleles) to contribute to developing depression

PKU

What is the diathesis stress model?

Shows how individuals inherit different vulnerabilities (diathesis) to develop psychopathology, but psychopathology only results when diathesis is triggered by stressful environmental factors

MAOA enzyme

breaks down serotonin, norepinephrine, and dopamine, reducing availability of neurotransmitters in synaptic cleft

MAOA and antisocial behavior in child maltreatment

Low MAOA activity increases neurotransmitters and child becomes more sensitive to maltreatment, which increases risk for antisocial behaviors

Limitations of the study: tiny sample size with small effect sizes

Gene environment correlation

when genetic differences are associated with systematic differences in environmental exposure; genotypes influence the kinds of environments experienced

Passive GxE correlation

When parents who transmit genes that promote the development of a specific trait are likely to also create a rearing environment that fosters the development of that trait (ex. An educated parent passes on intelligence, but also creates an environment where there are a lot of books, homework help, etc. Musically gifted parents transmit genes of musical ability to children and also have instruments, musical toys, lessons available for the child).

Reactive/evocative GxE correlation

Our experiences are in part a function of the reactions our behaviors elicit from others, and the extent our behavior is genetically influenced, this induces reactive (or evocative) GxE correlation (ex. a happy baby vs. a colic baby have different experiences based on how their behaviors elicit responses from their parents)

Active GxE correlation

Our environment is in part a function of the choices we make based upon our abilities and interests and the extent to which our abilities and interests are genetically influenced induces a GxE correlation (ex. Extroversion is heritable, and extroverted people choose different environments, like the party scene)

What is the impact of being reared together or apart on the correlation between personality scores of relatives?

Being reared together doesn’t lead to greater similarity in personality. Reared together and reared apart MZ twins are about as similar in personality, suggesting little environmental effect. MZ correlations are twice the DZ correlations, indicating a strong genetic effect on personality.

Wilson effect

Shared environmental influences decline over time while genetic influences increase over time.

Ex. IQ has a greater genetic effect in adulthood, whereas IQ has a greater environmental effect during childhood.

What other traits show the Wilson effect?

Cognitive ability, social values, cigarette use

How do the three laws of behavioral genetics explain individual differences?

1. All human behavioral traits are heritable (40-50% of individual differences)

2. The shared family environment has minimal impact on individual differences in behavior (0-10%)

3. The non-shared environment exerts a major influence on individual differences in behavior (50%)

What is positional cloning?

A method of gene identification in which a gene for a specific phenotype is identified only by its approximate chromosomal location (but not function); identify genetic variants contributing to multifactorial risk 

What is allelic association?

Population level association between allele status and phenotype in case-control or population studies

What statistical measure represents risk in allelic assocation

Odds ratio, which can achieve high statistical power to test for small effects by increasing sample size

What is the candidate gene approach to identifying genetic variants?

Sanders et al. found 14 candidate genes with multiple SNP variants for each gene, but only a few were significant. They also didn’t control for the total number of independent tests

What is SNP tagging?

SNP1 can be used to predict SNP2 (no use in genotyping both)

What is linkage disequilibrium?

non random population association between alleles linked loci (associations between non-functional genetic markers and phenotypes)

LD arises due to new mutations that arise on a haplotype background.

What is a haplotype?

alleles at different loci inherited together on the same chromosome

How does LD related to meiosis/recombination, number of generations, and physical distance in the genome?

During meiosis, homologous chromosomes exchange segments through recombination. Each recombination event can break the association between alleles at different loci.

The number of generations affects LD because recombination accumulates over time

The longer a population has existed since a mutation first arose, the more opportunities recombination has had to erode the association between that mutation and its original haplotype.

Recombination events (due to distance or many generations) break LD down.

What is a GWAS?

sequencing the entire genome to detect common variants associated with multifactorial disease

What is genetic architecture?

characterizing the specific genetic variants that influence complex (multifactorial) phenotypes in terms of…

the number of contributing genetic variants

their frequency of occurrence in the population

the magnitude of the effects on the phenotype

the mechanisms by which they affect risk

CDCV

cumulative effect of common variants of small effect, high allele frequency

common disorders are inherited due to aggregated effects of common variants

CDRV

multiple rare variants of large effect, low allele frequency

common disorders are heterogenous and each case has a different rare risk variant

Do common variants affect gene expression?

Yes, all phenotypes studied in GWAS had identified 50% of SNPs in introns that are spliced out. Only 4% of SNPs actually change protein sequence

Do rare variants affect gene expression?

No they affect gene product

exome sequencing

sequencing the coding regions of all protein-coding genes

What is missing heritability?

the proportion of phenotypic variance accounted for by GWAS is typically only a small portion of the biometric heritability

What could account for missing heritability?

Heritability estimates of twin studies are inflated

Non additive genetic effects (only additive genetic effects are added)

GWAS samples are not large enough to detect small SNP effects

GWAS detects common SNP variants but maybe rare variants contribute to missing heritability

what is a polygenic risk score?

a weighted, linear, combination of SNPs identified in a large GWAs as being associated with disease risk 

SNP heritability

mathematical estimate of the aggregate effect of all SNPs in the analysis

pleiotropy

one gene can have multiple phenotypic effects

genetic correlation

quantifies degree of genetic overlap for multifactorial phenotypes

population stratification

frequency of alleles varies across populations

How to address population stratifcation?

Including ancestral covariates in GWAS analyses to statistically control for subtle population structure

Using mixed models or linear mixed effects models that account for relatedness and population structure directly through a genomic relationship matrix

Restricting analyses to relatively homogeneous ancestry groups to reduce confounding, though this limits generalizability

Applying within-family designs, which automatically control for ancestry because family members share the same ancestral background

What is ancestry?

the region of the world where our ancestors came from (the people we are more genetically similar to)

Genetic drift

differences in allele frequency due to drawing a sample from a larger population (due to chance)

founder effects

taking a subset from a larger population, establishing a new population, reducing genetic variability

bottleneck effects

any random, external event that reduces population size and forms a new population, reducing genetic variability