Genetics - Mapping Common Variants

Front 1

Common variants

Back 1

Can't be genotyped yet, but know they exist due to reasonable evidence
- Growing body of knowledge

Front 2

Rare vs. Common variants

Back 2

- Rare variants = less common, huge risk factor for disease
- Common variants - Relatively common, presence adds little risk for disease

Front 3

Complex disease

Back 3

Results from combo of genes and environment
- Same as "multifactorial" or "polygenic" disease

Front 4

To confirm complex disease actually has genetic component

Back 4

Familial aggregation - families with common genes have larger incidence of given condition
- Twin studies confirm this, because family might all simply be exposed to same environmental factor
- Living next door to nuke power plant, etc.

Front 5

Twin studies

Back 5

Confirms that disease has genetic basis
- Monozygotic twins = identical genome
- Dizygotic twins (fraternal) = 1/2 of same genes
- Thus, if MZ concordance > DZ concordance = probably some complex/multifactorial disease present

Front 6

Prediction of common disease

Back 6

Currently, we don't really know many specific variants
- Thus, assume no knowledge of genetic risk factors for an individual
- Use info from "non-genetic" factors to determine risk

Front 7

Risk curve

Back 7

Bell-curve
- Past certain point on the right = going to get disease
- Absolute risk point doesn't change, but our curve shifts left/right based on factors

Front 8

Multifactorial liability threshold model

Back 8

Only those who have inherited enough susceptibility alleles will get disease

Front 9

Gender risk factors

Back 9

Some diseases show biases towards gender
- Higher-freq affected gender needs LESS susceptibiliy alleles to manifest disease
- Lower-freq affected gender needs MORE susceptibiliy alleles to manifest disease

Front 10

Pyloric stenosis

Back 10

5x more common in males
- Affected females - more likely to have affected children than affected males

Front 11

Pyloric stenosis risk

Back 11

Threshold farther to right for females than males - need more alleles to get disease
- Risk for relatives of affected MALE - have curve shifted to right, still more for males
- Risk for " of affected FEMALES - curve shifted even farther to the right, still gender bias

Front 12

Degree of relationship

Back 12

1st degree = share 1/2 genes
- 2nd degree = share 1/4 genes
- 3rd degree = share 1/8 genes

Front 13

Risk & Degree of relationship

Back 13

also a bell curve
- Risk for relatives of affected person - bell curve shifted to right
- Significant for 1st, continues to decline for 2nd, 3rd almost same as unrelated person

Front 14

Other risk factors

Back 14

1) High severity of disease (cleft lip/palate) - probably more alleles make more severe, higher likelihood of relatives inheriting
2) # of affected individuals - if lots of offspring being affected, probably greater risk

Front 15

Genome-wide association studies (GWAS)

Back 15

Comparing disease cases vs. control cases
- Compare freq. of SNP alleles, genotypes in disease and control cases
- Bottom line: is the frequency of any particular gene(s) higher in disease vs. control?
- Typically can only detect <20% of heritibility factors

Front 16

GWAS goal

Back 16

Personalized medicine - understanding individual needs, factors, dosages, risks, etc.

Front 17

GWAS of AD

Back 17

- Controls - older people, guaranteed don't have AD
- Disease - patients diagnosed with AD
- Not looking for causative mutations
- Looking for common markers to ID likely spots for contributing gene loci

Front 18

GWAS benefits

Back 18

- ID'd novel loci contributing to AD risk
- However, individual risk of each loci = very small...

Front 19

GWAS problems

Back 19

- Easily detects large effects in small/large sample populations, but really need very large sample sizes to isolate small effects...
- Good at detecting common variants - the more variants, the weaker the correlations
- If each disease sub-family has own variants, GWAS is useless