Behavioral Genetics [Guest Lecture]

Twin studies

• Rely on the fact that identical twins are on average twice more genetically similar to each other than fraternal twins

• Allow to estimate the heritability (H2) of a trait: what part of the inter-individual variance in the trait can be attributed to genetic factors.

The first law of Behavioral Genetics

all human behavioral traits are heritable (partly be genes)

• identical twins raised apart (usually highly) similar in every conceivable measurement

• behavioral similarity predicted by genetic similarity:

  • identical twins more similar than fraternal twins

  • exceptions: religion, language

the second law of behavioral genetics

effect of being raised in same family is smaller than the effects of the genes

• identical twins raised apart are almost as similar as identical twins rasied together

• non-related individuals reared together are almost as similar as random strangers

expensive schools, nice homes, strict discipline, religious background matter little for many traits relative to genetic factors

thirs law of behavioral genetics

substantial portion of the variation in complex human behavioral traits is not accounted for by the effects of genes or families

• identical twins (even raised tg) in fact far from identical and differ in significant ways

• variance left over once genes and shared environment are accounted for

human genome

sequence of about 3 billion base pairs that are indexed based on their location on the sequence

• inclues about 20k-30k genes: subsequences that contain biological info involved many biological processes

chromosome: genes packages into 23 structures

includes subsequences that are not genes

Single Nucleotide Polymorphisms (SNPs)

the most common type of genetic variation, representing differences in a single DNA building block (nucleotide) that occur in at least 1% of the population

genetic variation data

array of the bases in the most common SNP locations

not all genetic variants are SNPs

Insertions and Deletions (Indels)

• Small DNA segments (1-50 base pairs) that are inserted or deleted.

Copy Number Variations (CNVs)

Large segments (50+ base pairs) that are duplicated or deleted in the genome.

Structural Variants (SVs)

• Large-scale changes affecting chromosome structure (1,000+ base pairs).

• Includes deletions, duplications, inversions, and translocations.

Short Tandem Repeats (STRs)

• Repeated short DNA sequences (2-6 base pairs in length).

Variable Number Tandem Repeats (VNTRs)

• Longer repeated DNA sequences (10-100 base pairs).

Epigenetic Modifications (Not DNA changes, but affect expression)

candidate gene studies

theoretically motivated studies that tested specific associations between traits and SNPs in genes that we a priori hypothesized to be related to them

direct to costumer (DTC) genetic testing

appeared in early 2000s

• allows consumers to obtain genetic data without physician

• genome is sequenced by private company that retains data

• provides ancestry or health info

genome wide association studies (GWAS)

large scale data driven investigations of associations between complex traits and SNPS across entire genome

• due to large # of associations studied in GWAS = methodology emphasizes very conservative correction for multiple hypotheses testing, pre registration of analyses plan and replication in independent samples

continuting development of genotyping techniques tg with massive data collection efforts = led to rapid incline in samples

increase in sample size increased statistical power → identification of many associations between SNP and complex behavioral traits

why do candidate gene studies fail to replicate?

• small non representative samples

• no control for many important covariates (ses and height)

• researcher degree of freedom

• publication bias

fourth law of behavioral genetics

typical human behavioral trait associated with very many genetic variants, each of which accounts for very small percentage of behavioral variability

• mechanisms of association between most SNPs and behavioral traits are unknown (active field of study)

polygenic risk scores (PRS)

aggregates effects of many SNPs (often linear combination of the most significant SNPs identified in its GWAS)

• becomes increasingly accurate as sampel size increase

• using PRS to control for genetics when studying environment, causal inference, taste preferences

uniqueness of genetic data

• immutable: can be measured early in life, informative of future

• often informative about things that are unobservable

• correlates w almost everything one can think of

• informative of one’s relative

advantage of biological big data

• more representative

• statistical power to detect and accurately estimate small effect

• capacity to see nuanced, dose dependent effects

• controlling for obvious confounds

• investigating gene brain behavior paths

risky behavior

• negative association between grey matter volume and risky behavior