MATLAB: How to get access to the support vectors calculated by the SVMTRAIN function in the Bioinformatics Toolbox

Bioinformatics Toolboxvector

I am using the SVMTRAIN function in my analysis, and I am interested in knowing the support vectors that are used within the function. I would like the SVMTRAIN function to output the support vectors.

Best Answer

This enhancement has been incorporated in Release 2006a (R2006a). For previous product releases, read below for any possible workarounds:

Documentation on the SVMTRAIN function is missing from the Bioinformatics Toolbox.

Here is the additional information on the SVMTRAIN function:

The structure returned by the SVMTRAIN function, SVMStruct, contains within it the support vectors used in the SVMTRAIN function. These can be accessed in the SupportVectors element of the SVMStruct, with a command such as the following:

SVMStruct.SupportVectors

Related Solutions

MATLAB: Svmtrain

Each data sample must belong to a pre-known group for the purpose of svmtrain. The grouping variable is a list of which group each sample belongs to.

Do you have 3 features or 400 features? Are your two data sets indicating exactly two different groups, or are there entries from multiple groups within a data set? If your data is already split by groups, then your group variable should be 1 for each member of the first data set and 2 for each member of the second data set.

MATLAB: Bioinformatics

The first input of multialign is a collection of sequences to be aligned, this can be:

1) an array of structures with the "Sequence" and "Header" fields e.g.

>> seqs(1).Header = 'm157k10aa';
>> seqs(1).Sequence = 'TYNYMRQLVVDVVITNHYSV';
>> seqs(2).Header = 'm157g1faa';
>> seqs(2).Sequence = 'FATYFSPGFDPAQARTFEVAHG';
>> seqs(3).Header = 'm157g1baa';
>> seqs(3).Sequence = 'SSDIHMPSQFFDNIVEKCSQLFWVL';

2) just a cell array of strings (incase you do not care about the header names), e.g.

>> seqs2{1} = 'TYNYMRQLVVDVVITNHYSV';
>> seqs2{2} = 'FATYFSPGFDPAQARTFEVAHG';
>> seqs2{3} = 'SSDIHMPSQFFDNIVEKCSQLFWVL';

Usually you do not build these collections manually, for example the function FASTAREAD will return an array of structures if one file contains all your sequences.

Note that in my examples I made the aminoacid sequences up, your strings do not look like real aa, they look more like the header.

It may also be possible that you have one sequence per FASTA file and you have already done somethingh similar to this:

>> m157k10aa = fastaread('m157k10aa.fa')

Then you only need to build an structure array from the individual structures:

>> seqs3 = [m157k10aa,m157g1faa,m157g1baa]

HTH Lucio

Best Answer

Related Solutions

MATLAB: Svmtrain

MATLAB: Bioinformatics

Related Question