Understanding Heritability

Goal of Quantitative Genetics

"extent to which observed differences among individuals are due to genetic difference of any sort and to environmental differences of any sort without specifying what the specific genes or environmental factors are" Part of this is to estimate heritability Heritability is the proportion of phenotypic variance that can be explained by genotypic variance, within a particular population.

Twin and Family Study Methods

A major method historically involves observing familial resemblance Methods for studying human variation include the following types family studies, where sources of resemblance (i.e., genetic or environmental) vary across pair groups Adoption Designs Parent Child Comparisons Sibling Comparisons Twins Reared Apart Rare, but sensational Twins share one source of resemblance (i.e., genetic similarity only) Twins Reared Together Classic Twin Design Twins share two sources of resemblance (i.e., genetic and environmental similarity)

Heritability is an often misinterpreted statistic, but still can be useful

Again, heritability is the proportion of phenotypic variance that can be explained by genotypic variance, within a particular population.

Types of traits

Dichotomous/ Discrete or Categorical Categorical variables are those that have discrete categories or levels. Categorical variables can be further defined as dichotomous (two types, such as a dichotomy of having a syndrome or not). Quantitative/ Continuous / or Dimensional A continuous trait displays a range of expression (such as weight, height, etc.) rather than an all-or-none dichotomy

Types of traits- different ways to measure similarity

Dichotomous/ Discrete or Categorical For a dichotomous or categorical trait, pair similarity is measured using pairwise concordance rate Quantitative/ Continuous / or Dimensional For continuous trait, pair similarity is measured using a correlation (r): number of drinks in a four hour window ( can be 0 - whatever ) continuous

So why is heritability a useful construct?

Differences in heritability across populations are informative Are there genetic or environmental differences across these populations? For example: IQ is low SES groups may be largely explained by environmental variance (e.g., range of quality of schools, exposures to trauma, poor nutrition) compared to high SES groups where genetic differences are not masked by harmful environmental risk factors Differences in heritability across contexts are informative Are there differences across these context that limit environmental variance (i.e., historical trends, changes in policies, cultural restrictions/ freedoms, etc.)? For example: Heritability of height was lower in cohorts with less access to nutrition, whereas in more recent years proportionally more of the differences in height can be explained by differences in genes across individuals Does a phenotype in this context or population have a large or small genetic component? If you want to look for genes associated with a trait-- study in in a population/context where heritability is high

alcoholism

Discordant - one has it one doesn't Cordiant- if they both have it or they both do not

Heritability Equation

Heritability is a proportion of phenotypic variance, so Phenotypic variance (Vp) = 1 And phenotypic variance is explained entirely by genotypic variance (Vg) and/or environmental variance (Ve), so Phenotypic variance (Vp) = 1 = Vg + Ve Heritability is the proportion of phenotypic variance that can be explained by genotypic variance, within a particular population, so: Heritability (h2) = Vg / (Vg+Ve) Heritability (h2) = Vg/ Vp

Heritability Misconceptions Just because a trait is highly heritable, it does not mean that differences across populations are necessarily due to genes.

In this example figure, the heritability of plant height is 100% within each population (all of the differences are due to genetic differences in the plants, since the environment is perfectly controlled for). The main difference across populations is environmental (differences in soil nutrients) Therefore, even though plant height is highly heritable, we could not claim that the differences are due to genes or one population is "genetically superior."

Heritability Misconceptions Relatedly, if you possess a trait that is highly heritable, this is not necessarily fate or deterministic

It is still possible to change the outcome of this trait by manipulating the environment. An individual with a genetic propensity for high BMI (which is highly heritable) can still modify their outcome and lose weight through diet and exercise. It may be that the way to manipulate it environmentally (e.g., increase environmental variance and reduce heritability) has not been discovered yet! For example, a genetic trait with no known cure may soon be "alterable" through advances in gene therapy.

Correlation coefficient (r) is used to to measure how strong a relationship is between two continuous variables (i.e., score for individual in pair 1 vs. score for individual for pair 2).

MZ twins tend to be more similar, or strongly correlated than DZ twins

Calculating pairwise concordance

Pairwise concordance is calculated by the number of concordant pairs over the total number of pairs in the study (concordant + discordant pairs) Concordance means that both twins either have the trait or both do not, discordance are pairs where only one twin has shows the trait

Pairwise concordance example

Remember: concordance means that both twins either have alcoholism or both do not, discordance are pairs where only one twin has alcoholism What observations do you have about this figure?

What does it mean to say a phenotype is "heritable"?

Some proportion of phenotypic variance is explained by genotypic variance Heritability almost never is = 0, because Vg is almost never equal = 0. Turkheimer claimed in his paper "Three Laws of Behavior Genetics," that the first law of behavioral genetics is: All human behavioral traits are heritable. [That is, they are affected to some degree by genetic variation.]

Quantitative Genetics Pair Similarity Observations

The observation that genetically related pairs are more similar (more highly correlated or have higher concordance rates) than pairs that are less genetically similar suggests that genes play a role in the trait The observation that environmentally similar pairs are more similar (i.e., pairs reared in the same home) than pairs that are less environmentally similar suggests that environment plays a role in the trait For most traits, even MZ pairs are almost never perfectly identically (i.e., correlation does not equal 100 or concordance does not equal 1), meaning there are some environmental factors contributing to differences in the pair

Applying heritability

Therefore, it may not be meaningful to talk about whether behavioral traits is heritable, but rather the extent to which a behavioral trait is heritable in a particular context. Because the estimate is a function of the population, other important considerations about the population you are studying include How much phenotypic variance is there? How much environmental variance is there? What are other important characteristics of the population or context Heritability estimates can vary across...

Heritability Truth Heritability estimates how well we could predict a trait from knowing someone's genetic sequence

This is true, if we completely understood all the relevant genetic effects, but science is a far way off We are getting closer, particularly with certain medical or physical phenotypes FYI: there are huge areas science and agriculture trying to predict phenotypes such as milk yield through a through understanding of a cow's lineage, breeding value, or genetic sequence

Heritability is a property of the population not the individual

When the heritability of a trait is described, it reflects how much variability in the population is a consequence of genetic variance in that population. Heritability only describes sources of differences across people It does not "explain" why an individual has a trait- one cannot say that 50% of an individual's height is due to genes and 50% is due to their environment

Heritability Misconceptions Heritability does not mean a trait is immutable or fixed

While a trait may have a genetic component, this does not mean that environmental manipulation cannot affect it For example, the laws during Prohibition Era decreased the availability of alcohol differentially around the country. The trait may have been heritable, but the environment (e.g. policy) dramatically changed the nationals overall use of alcohol Similarly, this affected the nation's: (Vp) Phenotypic Variance (Vg) Environmental Variance and the heritability estimate (H2)!

genetic variance (or proportion of total variance explained by genetic variance)

is another way to say "heritability." SES- socioeconomic status where 0 is low and 100 is high.

Heritability

is the proportion of phenotypic variance that can be explained by genotypic variance, within a particular population.

Variance

refers to differences in the population (Vp) Phenotypic variance refers to differences in observable traits in the population (Vg) Genotypic variance refers to genetic differences (different combinations of allele) across individuals in a population (Ve) Environmental variance refers to different