1. About

Binomial Distribution (BD)

:

    The prior probability of k-tuple composition appears in one type of class can be calculated by:

qj = mj/M

where mj is the total number of k-tuple composition that appeared in the j-th type of class and M is the total occurrence frequency of all k-tuple composition in the benchmark dataset.
    The probability of the i-th k-tuple composition occurring in the j-th class of location can be calculated by the following formula:

p(nij)=∑(m=nij,Ni)(Ni!/m!(Ni-m))*(qj^m)*(1-qj)^(Ni-m)

where Ni is the total count of the i-th k-tuple composition in the benchmark dataset, nij represents the number of occurrences of the i-th nonamer in the j-th type of location, and the sum is taken from nij to Ni. The confidence level of the i-th k-tuple composition in the j-th class can be given by:

CLij=1-p(nij)



2. How to

Use

:

(1) Download the executable package BinomialDistribution.exe directly;

(2) Or you can use github to download them.

(3) Example to use.



3. The

Parameters

description:

    -k  [the number of kmer]
    -t  [type of sequence(require:DNA ,RNA or protein)]
    -f  [the path of sequencefile]

Links