NAME¶

SYNOPSIS¶

  use Bio::Tools::PSort::SVMLoc;

  # Load the model stored in the file 'sample.model' and Frequent Patterns in 'FreqPatt'
  $svm = new Bio::Tools::PSort::SVMLoc(Model => 'sample.model', FreqPattern => 'FreqPatt');

  # Classify a sequence in $seq, do not include FASTA headers!
  $svm->classify($seq);

DESCRIPTION¶

Bio::Tools::PSort::SVMLoc is a stripped down version of Algorithm::SVM design specifically for PSortb. If you need functionality such as training of SVM, please use that module instead.

CONSTRUCTOR¶

  # Load an existing SVM.
  $svm = new Bio::Tools::PSort::SVMLoc(Model  => 'sample.model', FreqPattern => 'FreqPatt');

A Bio::Tools::PSort::SVMLoc object must be created loaded from an existing model file.

An existing SVM is loaded from a file using the Model named parameter and the frequent patterns set. The model file should be of the format produced by the svm->train program (distributed with the libsvm library) or from the $svm->save() method in Algorithm::SVM.

METHODS¶

  $result = $svm->classify($seq);

The classify method is used to classify a set of data according to the loaded model. The method accepts a single parameter, which should be a protein sequence object. Returns a floating point number corresponding to the predicted value.

MAINTAINER¶

SEE ALSO¶

ACKNOWLEDGEMENTS¶