Pharmacogenomics Quiz 3 (Lec 9-11) Flashcards
In a contingency table, if we are trying to find the number of T alleles when T/T= 121 and T/C=297, what would the answer be? (T=?)
***121 x 2= 242
242 + 297= 539
What is the null hypothesis regarding the relationship between allele and drug resistance?
There is NO ASSOCIATION between any allele and drug resistance
(no effect of allele, want to prove this wrong)
What value of X^2 supports the null hypothesis?
A small value of X^2 supports the null
*a large value would reject the null (what we want to see)
What is a p value used for?
Determines if there is an effect or if the association occurred by chance
What does P > 0.1 mean?
No presumption against the null hypothesis
(there is no association, DO NOT REJECT THE NULL)
What does P < 0.05 mean?
Strong presumption against the null hypothesis
(significant association, reject the null, GOOD)
What does P < 0.01 mean?
Very strong presumption against the null hypothesis
(very significant association, definitely reject the null, VERY GOOD)
Does P measure the strength of an association relationship?
NO
-can be affected by other factors like sample size and allele frequency
What is an odds ratio?
A statistical measure of association between two events
-There is an increased risk for a phenotype in people who carry a specific genotype/allele compared to patients who do not carry it
How do we calculate an odds ratio?
Odds of phenotype in an individual with
the genotype/allele
———————————————————–
Odds of phenotype in an individual
without the genotype/allele
What is a hazards ratio (HR)?
Similar concept as odds ratio but mainly discussed in survival data
What does an Odds Ratio = 1 mean?
No association
What does an Odds Ratio > 1 mean?
Potentially increases risk
(allele is called “risk allele”)
**The greater the OR is, the higher risk the allele will confer to the phenotype
What does an Odds Ratio < 1 mean?
Potentially decreases the risk
(allele will be called “protective allele”)
**The smaller the OR is, the lower risk the allele will confer to the phenotype
What is the 95% Confidence Interval used to decipher?
Tells us over 95% probability that the association is confident
**is a statistical probability for OR (the standard error of OR)
What does a 95% CI > 1 mean?
There is significant Risk Effect
What does a 95% CI that contains 1 mean?
No statistical significance
What does a 95% CI < 1 mean?
There is significant protective effect
True or False: Due to the large number of tests required for many SNPs vs the single phenotype, there is a high probability that many SNPs will be associated with the phenotype just by chance
TRUE
*called a false positive
*results from multiple-testing
*the more SNPs tested, the higher the probability for a false positive
What value is used as a corrected significant GWAS P value?
(to limit false positives)
5x10^ (-8)
**used when testing a large number of SNPs such as in GWAS testing
What is the key to help find cause and effect while doing experiments to test a hypothesis?
CONTROL
-Positive control
-Negative control
What is negative control?
Conditions/media that do not produce a result
Ex: putting a bacteria in a culture media that you know won’t kill it, just want to make sure the bacteria grows
What is positive control?
The standard of care drug
In human clinical trials where having a negative control would be unethical, what do we do?
Compare to the standard of care
True or False: A large sample size is essential to have reliable results
True
Do clinical studies usually use median or mean?
Median
Why do clinical studies normally use median over mean?
-Faster
-Patient data may not be normally distributed
(only need 50% of patients to die to get median rate, if patients were to live a long time with the drug you would never get the mean)