Feb 17 lecture - Multiple Testing Normally, we expect a...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Multiple Testing ± Normally, we expect a p-value of 0.05 to occur 5% of the time by chance ± If we have done multiple tests, then our overall analysis may have many p<0.05 by chance ± If we did 100 independent tests in one experiment, then we expect five p<0.05 by chance in that one experiment ± For this kind of analysis, we expect a p=0.0005 to occur 5% of the time by chance ± How about when the tests are not independent??? ± Begin with the actual pedigree structures and phenotypes from the real data set ± Use a program to generate genotypes without regard to affection status ² May generate new artificial marker data based on real information about the markers (allele frequencies, position relative to other markers) ² Alternatively, may take the existing data and somehow shuffle it, so that the relationship between phenotype and genotype is broken Simulation Studies
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 ± Simulate as many markers as were actually analyzed one dataset with all these genotypes is called a replicate ± For each replicate, analyze the same way the
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/04/2011 for the course GENETICS 302 taught by Professor Hey during the Spring '11 term at Rutgers.

Page1 / 6

Feb 17 lecture - Multiple Testing Normally, we expect a...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online