GWAS Flashcards
looking at the threshold model for multifactorial disease: if individuals cross the threshold they will have what?
combination of increased number of risk alleles or increased number of environmental risk factors
the relative contribution of each locus to overall population depends on:
magnitude of the effect of the risk allele
allele frequency
dominance
epistasis
risk alleles may be
common but have a small effect
rare with a large effect
mixture of oligogenes modified by polygenes and modifiers
effect size determines ease of what
identification or mapping
only a small proportion of individuals with the at risk genotype may suffer from the disease T/F
T
there are only a small number of alleles that are associated with a large risk of a complex disease. which locus and mutation is associated with type 1 diabetes
HLA locus
DQβ1 asp57any
if individuals are homozygous for which allele they are mmore likely to get alzheimers
ApoE ε4
what is an SNP
variation in a single nucleotide that occurs at a specific position in the genome
if you compared any 2 genomes you would find a difference around ever how many bases
1kb
in a population comparison with multiple genomes you would expect a difference every…
300bp
how many common SNPs are there
15million
how do we measure the effect sizes of alleles
Odds ratio
how do we calculate the odds ration (OR)
(cases/non cases genotype1) / (cases/non cases genotype2)
an odds ratio less than 1 indicates
condition is less likely to occur with the the first genotype
an odds ratio of more than 1 indicated
condition is more likely to occur with the the first genotype
what is genotype relative risk (GRR)
the ratio of risk with the allele compared to without the allele
what is the difference between OR and GRR
Relative risk is a ratio of probabilities whereas odds ratio is a ratio of odds
OR can overestimate and magnify risk, especially when the disease is more common (eg, hypertension) and should be avoided in such cases if RR can be used. T/F
T
what are the steps for conducting a GWAS
- collect datasets of cases and controls
- genotype each subject with a genome wide panel of SNPs
- calculare the OR for each allele in the panel for each individual in the dataset
- for each allele in genome wide panel - is the OR significantly different in cases and controls
if the OR is significantly different in cases compared to controls you can conclude
such alleles are associated with the disease
what are the 2 reasons of why an allele can be associated with a disease
1) the allele causes the risk
2) the allele is correlated with the causative allele because of linkage disequilibrium
which is more common
1) the allele causes the risk
2) the allele is correlated with the causative allele because of linkage disequilibrium
more common to be in linkage disequilibrium
what are haplotype blocks
set of closely linked alleles/markers on a chromosome that, over evolutionary time, tend to be inherited together
haplotype blocks mean that we can use one tag and easily work out the intervening sequences T/F
T
how can we work out the intervening sequences
because of linkage disequilibrium in haplotype blocks
the HapMap project was an international consortium that did what
used commercial microarrays to map haplotype blocks in the human genome
what do you use the map of haplotype blocks for
to select panel of SNPs that effectively stand as proxy for all other SNPs across genome
in GWAS you can get type 1 false positives - what does this mean
each SNP as individual test p value of 0.05 means the result would occur less than 1 in 20 times by chance. as you test more SNPs there will be more false positives
to get genome wide significance you need a p value of equal to or less than what
10^-8
what is a type 2 false negative
set threshold too high (v small p value) and miss weak signals
what is the “missing heritability” problem
he fact that single genetic variations cannot account for much of the heritability of diseases, behaviors, and other phenotypes
where is the missing heritability
- moderately rare alleles with moderate effect
- polygenic model of inheritance (each allele has effect too small to be detected by GWAS)
- overestimation of additive heritability in twin studies
when the p value was relaxed in a GWAS for schizophrenia there were more alleles of significance what does that mean
evidence that statistics can cause genuine alleles ti fall short of significance in original study
what is pathway analysis
seeing if the hits make biologiccal sense eg genes coding for muscle in lung in respiratory diseases
if you look at possible number of alleles contributing to a disease and relax the p value what do you get
get more alleles
how has GWAS been useful
- has illuminated mechanisms (T1D)
- can be used to predict those at risk (BC)
- tells you about genetic architecture of common diseases and phenotypic traits
what causes type 1 diabetes
autoimmune destruction of insulin producing β cells in islets of Langerhan
how can you diagnose children with type 1 diabetes
autoantibodies are detectable before onset
how much heritability has been defined in type 1 diabetes
80%
50% of heritability in type one diabetes was associated with which allele
HLA
the other genes involved in heritability of type 1 diabetes were what kind of genes
immune related genes involved in antigen presentation
tolerance to insulin is established by neonatal tolerance to what
cytotoxic T cells that recognise insulin
if there is a cytotoxic T cell that recognises self (eg insulin) during neonatal period it is directed to which pathway
apoptopic
mutation in antigen presenting pathway - particularly in which genes cause the body to see insulin as foreign
BQb1 or IL2R
mutation in antigen presenting pathway causes which cells to be destroyed so no insulin is produced
pancreatic b-islet cells
breast cancer has how much lifetime risk for women
8-12%
name 2 diseases that are caused by alleles with a large risk
alzheimers
parkinsons
BRCA1/2 acoounts for how much of the heritability of breast cancer
5%
how much heritability does genes involved in DNA repair pathway acount for in breast cancer
15%
how much of the heritability of breast cancer comes from common alleles
40%
common alleles have a minor allele frequency of more than…
0.5
average risk age for breast cancer is what
50
what is the downside for screening for breast cancer early
- there may be a risk of overtreatment and intervention
- abnormality may not have caused any problems if left untreated
UK women are invited to mammography screeing at what age
47
what are non modifiable risk factors
polygenic risk score, family history, age of menarche, age of first child, age of menopause
what are modifiable risk factors
BMI, HRT and smoking
20% of preventable cancer would be avoided if the top 10% at risk adopted a healthy lifestyle T/F
T
women in the top decile of non modifiable risk factors would have the same risk as an average women if they decreased modifiable risk factors T.F
T
what is the interstitial model proposed by R.A fisher
As the number of genes grows very large, the
contribution of each gene becomes correspondingly smaller,
what largely accounts for the missing heritability of many traits
SNPs with effect sizes well below the genome wide statistical significance
what contributes to the missing heritability for diseases with major fitness consequences such as autism and schizophrenia
rare variants with larger effect sizes
what is often thought of as the quintessential polygenic trait
height
most ….. windows in the genome include variants that affect height
`100kb
genetic contribution to disease is heavily concentrated in regions that are
transcribed or marked by active chromatin in relative tissues
broadly functional categories contribute more total heritabiltiy than do genes in apparently disease relevant functional categories T/F
T
for schizophrenia, chrohns and rheumatoid arthritis, where where the largest contributors to heritability found
in the broadest category genes - protein binding
what are core genes
genes that have biologically interpretable roles in a disease
give an example of a core gene
C4 on schizophrenia risk
cell regulatory networks are highly connected meaning that any expressed gene is likely to effect the regulation or function of
core genes
in summary what does the omnigenic model propose
since core genes are largely outnumbered by peripheral genes, a large fraction of the total genetic contribution to a disease comes from peripheral genes that do not play direct roles in the disease
In conclusion, our molecular genetic data strongly support a polygenic basis to schizophrenia that (1) involves common SNPs, (2) explains at least one-third of the total variation in liability, (3) is substantially shared with bipolar disorder, and (4) is largely not shared with several non-psychiatric diseases T/F
T
