Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
23 Cards in this Set
- Front
- Back
|
"Common"
|
1:3 to 1:100
|
|
|
"Intermediate"
|
1:100 to 1:1000
|
|
|
"Rare"
|
1:1000 to 1:1,000,000
|
|
|
Total nucleotide variants
|
3 million (of 6 billion NT)
1/3 of variants are in genes = 1 million |
|
|
Haplotype
|
Closely linked alleles inherited as a unit
Crossing over only occurs about 2x per chromosome Only a few of the possible haplotypes are actually observed |
|
|
Linkage disequilibrium
|
Blocks of genes passed down together
Breaks up more and more over generations |
|
|
Quantitative traits
|
Fall into a normal distribution: continuous distribution
Determines mean and outliers Affected/unaffected is discontinuous distribution yes or no |
|
|
Quantitative trait loci
|
genetic loci that affect the trait
individually are not predictive but constellation will be informative (e.g. blood pressure) |
|
|
Alzheimer's
|
75% is sporadic
25% familial but may not be Mendelian; indistinguishable clinically from sporadic ApoE: influences age of onset E3 polymorphism most common (cys/arg) E2 polymorphism (cys/cys) seems to generate some protection E4 polymorphism (arg/arg) increases odds ratio; but neither necessary, nor sufficient for Alzheimer's |
|
|
Case control study
|
Certain number of people with disease, certain number without
Looking for risk factor Looking for proportion that have the disease and the risk factor and the proportion of those without disease that have the risk factor (to show how predictive it is) |
|
|
HIV
|
Mutation in CCR5, delta 32, seem to be resistant to infection
|
|
|
Mendelian diseases: traits
|
Rare
Single gene affected High penetrance High genetic influence Low frequency alleles Tidy pedigrees Gene isolation relatively easy |
|
|
Multifactorial disease: traits
|
Common
Polygenic Environmental influence High frequency alleles Gene isolation difficult High public health impact |
|
|
Odds ratio
|
Affected with SNP/Affected without SNP
_________________________________ Unaffected with SNP/Unaffected without SNP (TL/BL over TR/BR) |
|
|
Relative risk
Risk ratio |
Risk of disease among exposed: Affected with SNP/Total with SNP
Risk of disease among unexposed: Affected without SNP/Total without SNP Exposed risk/unexposed risk = risk ratio |
|
|
P value
|
0.001 not meaningful when testing 1,000,000 markers --> 1,000 spurious (chance) associations would be observed
p = 3 X 10-8 is accepted |
|
|
GWAS weaknesses
|
Genome is large
P = 3 X 10-8 Role of any one genetic variant is likely to be small Statistical significance doesn't translate to clinical relevance Difficult to nail down a particular gene |
|
|
Weaknesses risk prediction
|
Relies on binary of affected, unaffected
Imprecise diagnostic criteria Misclassification of those are are at risk as not at risk False positives and false negatives |
|
|
Sensitivity
|
Proportion of those with the disease that are correctly identified by the test
TP/TP + FN |
|
|
Specificity
|
Proportion of those without disease that are correctly identified by the test
TN/TN + FP |
|
|
Positive predictive value (PPV)
|
Proportion of those with a positive test result that are correctly diagnosed as affected
TP/TP + FP |
|
|
Negative predictive value (NPV)
|
Proportion of those with negative test result that are correctly diagnosed as being unaffected
TN/TN + FN |
|
|
ROC curve
|
Farther away from dashed line, the more discerning the test
Can tune tests to be more sensitive or more specific |
