# source:orange-bioinformatics/docs/reference/obiProb.htm@1634:32373d957326

Revision 1634:32373d957326, 2.8 KB checked in by mitar, 2 years ago (diff)

Line
1<html>
2
4<title>obiProb: Probability distributions and corrections</title>
8
9<body>
10<h1>obiProb Probability distributions and corrections</h1>
11<p>obiProb provides the functionality to calculate probability distributions and corrections for multiple hypothesis testing.</p>
12
13<h2>Binomial</h2>
14<index name="Binomial">
15<p>A class for computing binomial distribution probabilities. <a href="http://en.wikipedia.org/wiki/Binomial_distribution">Binomial distribution</a> is the discrete probability distribution of the number of successes in a sequence of n independent yes/no experiments, each of which yields success with probability p</p>
16<p class=section>Methods</p>
17<dl class=attributes>
18    <dt>__call__(k, N, m, n)</dt>
19    <dd>if m out of N experiments are positive return the probability that k out of n experiments are positive using the binomial distribution.
20    (i.e. if p = m/N then return bin(n,k)*(p**k + (1-p)**(n-k)) where bin is the binomial coefficient)</dd>
21    <dt>p_value(k, N, m, n)</dt>
22    <dd>the probability that k or more tests are positive using the binomial distribution</dd>
23</dl>
24
25<h2>Hypergeometric</h2>
26<index name="Hypergeometric">
27<p>A class for computing hypergeometric distribution probabilities. <a href="http://en.wikipedia.org/wiki/Hypergeometric_distribution">Hypergeometric distribution</a> is a discrete probability distribution that describes the number of successes in a sequence of n draws from a finite population without replacement</p>
28<p class=section>Methods</p>
29<dl class=attributes>
30    <dt>__call__(k, N, m, n)</dt>
31    <dd>if m out of N experiments are positive return the probability that k out of n experiments are positive using the hypergeometric distribution.
32    (i.e. return bin(m, k)*bin(N-m, n-k)/bin(N,n) where bin is the binomial coefficient)</dd>
33    <dt>p_value(k, N, m, n)</dt>
34    <dd>the probability that k or more tests are positive using the hypergeometric distribution</dd>
35</dl>
36
37<h2>FDR</h2>
38<index name="FDR">
39<p>A function for preforming <a href="http://en.wikipedia.org/wiki/False_discovery_rate">False Discovery Rate</a> correction on a ordered list of p-values</p>
40<p class=section>Arguments</p>
41<dl class=attributes>
42    <dt>p_values</dt>
43    <dd>an ordered list of p-values</dd>
44    <dt>dependent (default False)</dd>
45    <dd>use correction for dependent hypotheses</dd>
46    <dt>m (default len(p_values))</dt>
47    <dd>number of hypotheses tested</dd>
48</dl>
49
50<h2>Bonferroni</h2>
51<index name="Bonferroni">
52<p>A function for performing <a href="http://en.wikipedia.org/wiki/Bonferroni_correction">Bonferroni correction</a> on a list of p-values</p>
53<p class=section>Arguments</p>
54<dl class=attributes>
55    <dt>p_values</dt>
56    <dd>a list of p-values</dd>
57    <dt>m (default len(p_values))</dt>
58    <dd>number of hypotheses tested</dd>
59</dl>
Note: See TracBrowser for help on using the repository browser.