inferential statistics Flashcards
Why normalise the data?
To control for variation in response due to factors not directly relevant to the action of the drug e.g. quality of the preparation. Variation receptor expression. Which might affect the magnitude of the response but not the shape of the concentration response curve
What does normalising allow?
To put data from different preps. Into comparable units. Which allows you to compare the shape or points on concentration response curves without being distracted by maximum or minimum response records
What is the EC50?
The molar concentration of an agonist that produces 50% of the maximal possible effect of that agonist. The action of the agonist may be stimulatory or inhibitory.
What is ED50?
the dose that produces 50% of the maximal response to that drug. ED50 is usually obtained from in vivo measurements which absolute drug concentration is uncertain.
What is pEC50?
The negative log of the molar concentration of an agonist that produces 50% of the maximal possible effect of that agonist.
What is a advantage of pEC50 over EC50?
Easier and clearer to view the data
Sigmoid function takes into account pEC50 not the EC50
What is wrong with doing a bunch of t-tests?
First one we get 95% certain of not committing type I error, but then we are only 90% certain then 80% certain. (0.95x2=0.90)
(0.95x3=0.86)
What is analysis of variance (ANOVA)?
Tests whether a variable significantly varies across three or more treamnets
What test do you conduct to see which group is different from the others?
Post- hoc tests (t-tests that modified) to keep you 95% that you are not committing a type 1 error.
What can Tukeys t be used for?
To make all-pairwise comparisons - compare one group against another
What is the Dennetts test?
Another post-hoc test ,it makes comparison relative to an identified control mean
