Lecture 2: Probability and population genetics Flashcards
Difference between random variable and realisation
A random variable has a list of outcomes. When an experiment has been performed and the variable taken a value this is termed a realisation.
What is a probability distribution?
Function that assigns a probability for a subset of outcomes for a random variable.
Name the most important discrete distributions
Bernulli
Binomial
Poisson
Name the most important continuous distributions:
The normal distribution
Exponential
chi-square
beta
Marginal density function
It is the a sub-density function. e.g. look only at the whole y distribution.
Conditional density function
keep at at certain value and look at the other distribution at that value.
What is independence?
If knowledge of X gives no information of Y.
F_{Y| X=x}(y) = F_Y(y)
Which gives:
F_{X,Y)(x,y) = F_X(x)*f_Y(y)
Covariance:
Expected value of the product of realisation minus expectation.
Independence implies
Uncorrelatedness
What is allele frequency estimation?
Estimation of the proportion of chromosomes that carry a specific allele.
can be done using number of people with genotype AA, Aa and aa (nAA, nAa and, naa respectively. Proportion of A is then:
p = (2nAA 0 nAa)/2n
a freq is 1-p
Variance of allele estimation:
The chromosomes in a sample can be thought of as independent. This means that the amount of alleles is binomially distributed with:
p = X (number of alleles) / 2n
2np ~ Binom(2n,p)
V(p) = p(1-p)/2n