01.04.2015 Views

Sequence Comparison.pdf

Sequence Comparison.pdf

Sequence Comparison.pdf

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

B.3 Major Discrete Distributions 179<br />

A random variable is discrete if it takes a value from a discrete set of numbers.<br />

For example, the number of heads turning up in the experiment of tossing a coin n<br />

times is a discrete random variable. In this case, the possible values of the random<br />

variableare0,1,2,...,n.Theprobability distribution of a random variable is often<br />

presented in the form of a table, a chart, or a mathematical function. The distribution<br />

function of a discrete random variable X is written F X (x) and defined as<br />

F X (x)=∑ Pr[X = v], −∞ < x < ∞.<br />

v≤x<br />

(B.6)<br />

Notice that F X is a step function for a discrete random variable X.<br />

A random variable is continuous if it takes any value from a continuous interval.<br />

The probability for a continuous random variable is not allocated to specific values,<br />

but rather to intervals of values. If there is a nonnegative function f (x) defined for<br />

−∞ < x < ∞ such that<br />

∫ b<br />

Pr[a < X ≤ b]= f (x)dx, ∞ < a < b < ∞,<br />

a<br />

then f (x) is called the probability density function of the random variable X. IfX<br />

has a probability density function f X (x), then its distribution function F X (x) can be<br />

written as<br />

∫ x<br />

F X (x)=Pr[X ≤ x]= f X (x), −∞ < x < ∞.<br />

(B.7)<br />

−∞<br />

B.3 Major Discrete Distributions<br />

In this section, we simply list the important discrete probability distributions that<br />

appear frequently in bioinformatics.<br />

B.3.1 Bernoulli Distribution<br />

A Bernoulli trial is a single experiment with two possible outcomes “success” and<br />

“failure.” The Bernoulli random variable X associated with a Bernoulli trial takes<br />

only two possible values 0 and 1 with the following probability distribution:<br />

Pr[X = x]=p x (1 − p) 1−x , x = 0,1,<br />

(B.8)<br />

where p is called the parameter of X.<br />

In probability theory, Bernoulli random variables occur often as indicators of<br />

events. The indicator I A of an event A is the random variable that has 1 if A occurs<br />

and 0 otherwise. The I A is a Bernoulli random variable with parameter Pr[A].

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!