testGeneFamilies.hypergeometric {ShotgunFunctionalizeR}R Documentation

Identified significant gene families between two samples using the hypergeometric model

Description

Test for significant gene families using the hypergeometric model.

Usage

testGeneFamilies.hypergeometric(data)

Arguments

data $2times n$ matrix containing counts

Details

testGeneFamilies.binomial performs an exact hypergeometric test for each row of in the data matrix. Under the assumption that the counts from both metagenomes follows a multinomial distribution, the procedure test whether $p_{i1}=p_{i2}$ for each row $i=1, ..., n$. If $X_{i1}$ and $X_{i2}$ are the counts for row $i$ and metagenome 1 and 2 respectively, it follows that [ X_{i1}|X_{i1}+X_{i2}=n sim Hyper(M_1+M_2, M_1, n) ] where $M_1$ and $M_2$ are the total number of counts in metagenome 1 and 2 respectively. The function testGeneFamilies.binomial provides a faster binomial approximation.

testGeneFamilies.hypergeometic is usually called implicitly by using the 'binomial' method in testGeneFamilies.

Value

testGeneFamilies.binomial returns a list with the following elements

PValues A list of p-values
Rankings.detail A list additional elements containing details about the ranking

See Also

testGeneFamilies, testGeneFamilies.binomial, testGeneFamilies.gaussian, testGeneFamilies.poisson


[Package ShotgunFunctionalizeR version 1.0-2 Index]