void RTSimulation::wrapSVM(std::vector<AASequence>& peptide_sequences, std::vector<DoubleReal>& predicted_retention_times)
{
String allowed_amino_acid_characters = "ACDEFGHIKLMNPQRSTVWY";
SVMWrapper svm;
LibSVMEncoder encoder;
svm_problem* training_data = NULL;
SVMData prediction_samples;
SVMData training_samples;
UInt k_mer_length = 0;
DoubleReal sigma = 0.0;
UInt border_length = 0;
Size max_number_of_peptides(2000); // hard coding pediction bins; larger values only take more memory, result is not affected
LOG_INFO << "Predicting RT ... ";
svm.loadModel(rt_model_file_);
// load additional parameters
if (svm.getIntParameter(SVMWrapper::KERNEL_TYPE) == SVMWrapper::OLIGO)
{
String add_paramfile = rt_model_file_ + "_additional_parameters";
if (!File::readable(add_paramfile))
{
throw Exception::InvalidParameter(__FILE__, __LINE__, __PRETTY_FUNCTION__, "RTSimulation: SVM parameter file " + add_paramfile + " is not readable");
}
Param additional_parameters;
ParamXMLFile paramFile;
paramFile.load(add_paramfile, additional_parameters);
if (additional_parameters.getValue("border_length") == DataValue::EMPTY
&& svm.getIntParameter(SVMWrapper::KERNEL_TYPE) == SVMWrapper::OLIGO)
{
throw Exception::InvalidParameter(__FILE__, __LINE__, __PRETTY_FUNCTION__, "RTSimulation: No border length defined in additional parameters file.");
}
border_length = ((String)additional_parameters.getValue("border_length")).toInt();
if (additional_parameters.getValue("k_mer_length") == DataValue::EMPTY
&& svm.getIntParameter(SVMWrapper::KERNEL_TYPE) == SVMWrapper::OLIGO)
{
throw Exception::InvalidParameter(__FILE__, __LINE__, __PRETTY_FUNCTION__, "RTSimulation: No k-mer length defined in additional parameters file.");
}
k_mer_length = ((String)additional_parameters.getValue("k_mer_length")).toInt();
if (additional_parameters.getValue("sigma") == DataValue::EMPTY
&& svm.getIntParameter(SVMWrapper::KERNEL_TYPE) == SVMWrapper::OLIGO)
{
throw Exception::InvalidParameter(__FILE__, __LINE__, __PRETTY_FUNCTION__, "RTSimulation: No sigma defined in additional parameters file.");
}
sigma = ((String)additional_parameters.getValue("sigma")).toFloat();
}
svm.setParameter(SVMWrapper::BORDER_LENGTH, (Int) border_length);
svm.setParameter(SVMWrapper::SIGMA, sigma);
// loading model data
String sample_file = rt_model_file_ + "_samples";
if (!File::readable(sample_file))
{
throw Exception::InvalidParameter(__FILE__, __LINE__, __PRETTY_FUNCTION__, "RTSimulation: SVM sample file " + sample_file + " is not readable");
}
training_samples.load(sample_file);
svm.setTrainingSample(training_samples);
svm.setTrainingSample(training_data);
// use maximally max_number_of_peptides peptide sequence at once
Size tmp_count = 0;
Size count = 0;
std::vector<AASequence>::iterator pep_iter_start = peptide_sequences.begin();
std::vector<AASequence>::iterator pep_iter_stop = peptide_sequences.begin();
while (count < peptide_sequences.size())
{
while (pep_iter_stop != peptide_sequences.end() && tmp_count < max_number_of_peptides)
{
++tmp_count;
++pep_iter_stop;
}
std::vector<AASequence> tmp_peptide_seqs;
tmp_peptide_seqs.insert(tmp_peptide_seqs.end(), pep_iter_start, pep_iter_stop);
std::vector<DoubleReal> tmp_rts(tmp_peptide_seqs.size(), 0);
std::vector<DoubleReal> tmp_pred_rts;
// Encoding test data
encoder.encodeProblemWithOligoBorderVectors(tmp_peptide_seqs, k_mer_length, allowed_amino_acid_characters, border_length, prediction_samples.sequences);
prediction_samples.labels = tmp_rts;
svm.predict(prediction_samples, tmp_pred_rts);
predicted_retention_times.insert(predicted_retention_times.end(), tmp_pred_rts.begin(), tmp_pred_rts.end());
pep_iter_start = pep_iter_stop;
count += tmp_count;
tmp_count = 0;
}
LibSVMEncoder::destroyProblem(training_data);
LOG_INFO << "done" << endl;
}
