Ecotilling Marilia Lobo Burle ECL 290 Winter 2003 - PowerPoint PPT Presentation

1 / 7
About This Presentation
Title:

Ecotilling Marilia Lobo Burle ECL 290 Winter 2003

Description:

Equation: is the difference between the observed and expected haplotypes frequencies. ... PAB are haplotype freq. Other statistics for LD. Dab = (PAB - PAPB) ... – PowerPoint PPT presentation

Number of Views:181
Avg rating:3.0/5.0
Slides: 8
Provided by: Kami86
Category:

less

Transcript and Presenter's Notes

Title: Ecotilling Marilia Lobo Burle ECL 290 Winter 2003


1
Linkage disequilibrium Nonrandom association of
alleles at different loci Causes shared history
of mutation and recombination LD gametic phase
disequilibrium gametic disequilibrium, allelic
association LD confused with linkage linkage
physical connection in a chromosome LD
correlation between alleles in a population L
Equilibrium large randomly mated population no
selection, mutation, migration. LD linkage,
selection and admixture. LD decays over time
with random mating and recombination, but
eventually slowly
2
Calculation Locus 1 with alleles A and a Locus
2 with alleles B and b D (pAB pA.p.B)
PA is allele freq. PAB are haplotype freq.
Equation is the difference between the observed
and expected haplotypes frequencies. Perform
statistical tests
3
Other statistics for LD Dab (PAB - PAPB)   r2
? 2 (Dab)2 / PA Pa PB Pc r2 summarizes
both recombinational and mutational
history D (Dab)2 / min (PA Pb, Pa PB)
for Dab lt 0   D (Dab)2 / min (PA PB, Pa
Pb) for Dab gt 0 D measures only
recombinational history
4
Flint-Garcia et al., 2003
5
Causes of LD -LD in the founding
population -admixture of populations with
different gametic frequencies -selection
favoring some genotypes over others. How
programs perform test If a locus NOT in H-W,
use genotype frequency instead of allele freq. to
LD test. GDA a contingency table for all pairs
measure D tests significance may use genotypes
if locus not H-W Popgene Burrows composite
measure of LD, ?2 test. Genepop a contingency
table for all pairs genotypic LD and gametic LD
tests significance.
6
Bonferroni correction Reduces a (type I
error) Type I error incorrectly rejecting Ho
(incorrectly declaring a difference). When use?
more tests independent Ex. There is no real
difference 6 tests a 0.05 prob. NOT getting
differences 0.95 6 tests (0.95)6 0.74 a
0.05/K K number of tests. Increases ß (type
II error) Type II error incorrectly accepting
Ho (incorrectly declaring no difference)
7
When Why Do You Want to Use Allelic Vs.
Genotypic Frequencies for LD Calculations?
Locus in a population
  • In HWE
  • Allelic frequencies should be used to calculate
    LD
  • Not in HWE (has within-locus disequilibrium)
  • Pairwise disequilibrium includes all types of
    disequilibrium
  • (within-locus between-loci)
  • Genotypic frequencies should be used to
    calculate LD to
  • prevent interference from within-locus
    disequilibrium
Write a Comment
User Comments (0)
About PowerShow.com