///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void EnergyPlot::run() {
/*
* Make sure the dot plot data can be accessed.
*/
// Print a message that says the dot plot data file is being checked.
cout << "Checking dot plot input file..." << flush;
// Initialize the dot plot handler and an error string.
DotPlotHandler* plotHandler = 0;
string error = "";
// Create the RNA strand that holds the dot plot data.
RNA* strand = new RNA( inputFile.c_str(), 4 );
ErrorChecker<RNA>* checker = new ErrorChecker<RNA>( strand );
error = checker->returnError();
if( error == "" ) { cout << "done." << endl; }
/*
* Prepare the dot plot data.
*/
if( error == "" ) {
// Print a message saying that the dot plot data is being prepared.
cout << "Preparing dot plot data..." << flush;
// Build the dot plot handler using the strand's sequence length.
int length = strand->GetSequenceLength();
plotHandler = new DotPlotHandler( descriptionSettings.getDescription(inputFile, true), length );
// Add all possible dots to the dot plot.
for( int i = 1; i <= length; i++ ) {
for( int j = i; j <= length; j++ ) {
double energy = strand->GetPairEnergy( i, j );
if( energy > 0.0 ) { energy = numeric_limits<double>::infinity(); }
else {
stringstream stream( stringstream::in | stringstream::out );
stream << fixed << setprecision( 1 ) << energy;
stream >> energy;
}
plotHandler->addDotValue( i, j, energy );
}
}
// Print a message saying that the dot plot data has been prepared.
cout << "done." << endl;
}
//fetch each attribute, error checking, build rowobject
//input: string of row from csv, colindex for data type
Teach_rowObject TeachingRowBuilder::buildRow(string data, ColIndex index){
//*** instantiate classes that will help out in this function
AttributeRetriever fetch(data);
bool hasError = 0; //Set this to 1 if an error is found
ErrorChecker filter;
bool enableErrorChecking (1);
//*** retrieve each attribute - temporarily store them
string name = fetch.getAttribute(index.name_loc);
string domain = fetch.getAttribute(index.domain_loc);
string program = fetch.getAttribute(index.program_loc);
string courseType = fetch.getAttribute(index.courseType_loc);
string geoScope = fetch.getAttribute(index.geoScope_loc);
string title = fetch.getAttribute(index.title_loc);
int sDate = fetch.getIntAttribute(index.sDate_loc);
int eDate = fetch.getIntAttribute(index.eDate_loc);
int nTeach = fetch.getIntAttribute(index.nTeach_loc);
int tStudents = fetch.getIntAttribute(index.tStudents_loc);
float hpTeach = fetch.getIntAttribute(index.hpTeach_loc);
float tHours = fetch.getIntAttribute(index.tHours_loc);
//*** check for errors
if (enableErrorChecking){
//strings - check for blank entries
name = filter.blankCatch(name, hasError);
domain = filter.blankCatch(domain, hasError);
program = filter.blankCatch(program, hasError);
courseType = filter.blankCatch(courseType, hasError);
//dates, check for zeroes
sDate = filter.zeroCatch(sDate, hasError);
eDate = filter.zeroCatch(eDate, hasError);
hpTeach = filter.zeroCatch(hpTeach, hasError);
tHours = filter.zeroCatch(tHours, hasError);
}
//*** Build Row
Teach_rowObject currentRow (hasError, name, domain, program, courseType, geoScope, nTeach, sDate, eDate, hpTeach, tHours, tStudents);
return currentRow;
};
///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void efn2Interface::run() {
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for RNA that specifies a filename.
* Specify type = 1 (CT file).
* isRNA identifies whether the strand is RNA (true) or DNA (false).
*
* After construction of the strand data structure, create the error checker which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant of the isErrorStatus method, which returns 0 if no error occurred.
* The calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
RNA* strand = new RNA( ctFile.c_str(), 1, isRNA );
ErrorChecker<RNA>* checker = new ErrorChecker<RNA>( strand );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Check the strand for pseudoknots by looking through each structure with the ContainsPseudoknot method.
* efn2 cannot handle pseudoknots, so if the strand contains pseudoknots, this is considered an error.
*/
if( error == 0 ) {
structures = strand->GetStructureNumber();
for( int i = 1; i <= structures; i++ ) {
if( strand->ContainsPseudoknot( i ) ) {
cerr << "Nucleic acids contain pseudoknots; cannot proceed." << endl;
error = -1;
}
}
}
/*
* Set the temperature using the SetTemperature method.
* Only set the temperature if a given temperature doesn't equal the default.
* If the temperature does need to be set, use the error checker's isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && ( temperature != 310.15 ) ) {
// Show a message saying that the temperature is being set.
cout << "Setting temperature..." << flush;
// Set the temperature and check for errors.
int tempError = strand->SetTemperature( temperature );
error = checker->isErrorStatus( tempError );
// If no error occurred, print a message saying that temperature is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Read SHAPE constraints, if applicable.
* When reading SHAPE constraints, use the ReadSHAPE method.
* After constraints are read, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 && SHAPEFile != "" ) {
// Show a message saying that SHAPE constraints are being read.
cout << "Applying SHAPE constraints..." << flush;
// Initialize the single stranded SHAPE slope and intercept.
// For now, these are hard-coded as 0.
double slopeSingle = 0;
double interceptSingle = 0;
// Read SHAPE constraints and check for errors.
int constraintError = strand->ReadSHAPE( SHAPEFile.c_str(), slope, intercept, slopeSingle, interceptSingle );
error = checker->isErrorStatus( constraintError );
// If no error occurred, print a message saying that SHAPE constraints are set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Do the efn2 calculation.
* If the user wants a simple output file, get free energies for each structure using the CalculateFreeEnergy method.
* If the user wants a thermodynamic details file, write the file with the WriteThemodynamicDetails method.
*/
if( error == 0 ) {
// Write a thermodynamic details file, if asked.
if( writeTherm ) {
// Show a message saying that the details file is being written.
cout << "Writing thermodynamic details file..." << flush;
// Write the thermodynamic details file and check for errors.
int thermError = strand->WriteThermodynamicDetails( outFile.c_str() );
error = checker->isErrorStatus( thermError );
// Print a message saying that the details file has been written.
if( error == 0 ) { cout << "done." << endl; }
}
// Write a simple list file, if asked.
else {
// Show a message saying that the list file is being written.
cout << "Writing free energy list file..." << flush;
// For each structure, calculate its energy and push it into the energies vector.
// Changed this 5-14-2013, no longer pushes into a vector. Instead use calculatefreeenergy method from RNA class.
// This is so it works with openMP -- vector would be filled out of order when loop runs in parallel and the indexing would be messed up.
// If an error occurs, set it.
//vector<double> energies;
strand->CalculateFreeEnergy( 1 , simple ); //calculate the free energy of the first structure first (this is to make it play well with openMP)
#ifdef SMP
#pragma omp parallel for
#endif
for( int i = 2; i <= structures; i++ ) {
strand->CalculateFreeEnergy( i , simple ); //now calculate the free energy of the remaining structures
error = checker->isErrorStatus();
//if( error == 0 ) { energies.push_back( energy ); }
//else break;
}
// If all free energies were calculated correctly, write the output file.
if( error == 0 ) {
ofstream out( outFile.c_str() );
for( int i = 1; i <= structures; i++ ) {
int index = i - 1;
out << "Structure: " << i << " Energy = " << fixed << setprecision( 1 ) << strand->GetFreeEnergy(i) << endl; //changed to GetFreeEnergy instead of writing the energies vector
}
out.close();
}
// Print a message saying that the list file has been written.
if( error == 0 ) { cout << "done." << endl; }
// If the output should be piped to standard output, then pipe it.
if( ( error == 0 ) && ( stdPrint ) ) {
cout << endl << "Generated output file: " << outFile << endl << endl;
for( int i = 1; i <= structures; i++ ) {
cout << "Structure: " << i << " Energy = " << fixed << setprecision( 1 ) << strand->GetFreeEnergy(i) << endl;
}
cout << endl;
}
}
}
// Delete the error checker and data structure.
delete checker;
delete strand;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
///////////////////////////////////////////////////////////////////////////////
// Read base pair probabilities from a partition function save file.
///////////////////////////////////////////////////////////////////////////////
void Postscript_Annotation_Handler::readPartition( string file,
RNA* structureStrand ) {
// Initialize the RNA strand and error checker that reads partition data.
RNA* partStrand = new RNA( file.c_str(), PFS_TYPE );
ErrorChecker<RNA>* partChecker = new ErrorChecker<RNA>( partStrand );
// If the RNA strand and error checker were created successfully, read in the
// annotation data.
if( !( error = partChecker->isErrorStatus() ) ) {
// If there are no structures in the strand, print out an error message.
// Otherwise, initialize the annotation array to handle the appropriate
// amount of structures.
if( structures == 0 ) {
cerr << "No structures or pairs are present to annotate." << endl;
error = true;
} else {
probabilityAnnotations.resize( structures );
for( int i = 1; i <= structures; i++ ) {
vector<char> row;
row.resize( length );
for( int j = 1; j <= length; j++ ) { row[j-1] = 'i'; }
probabilityAnnotations[i - 1] = row;
}
}
// For each structure possible, read in its base pair probability data.
for( int i = 1; i <= structures; i++ ) {
// If an error has occurred, stop reading data.
if( error ) { break; }
// Loop through the structure to find pairs.
for( int j = 1; j <= length; j++ ) {
// If an error has occurred, stop reading data.
if( error ) { break; }
// Get the next pair. If an error occurred, stop reading data.
int pair = structureStrand->GetPair( j, i );
int code = structureStrand->GetErrorCode();
if( code != 0 ) {
cerr << endl << structureStrand->GetErrorMessage( code ) << endl;
error = true;
break;
}
// If the next nucleotide is in fact paired, determine the proper
// color code for it.
if( ( pair != 0 ) && ( pair > j ) ) {
// Get the probability for this pair.
// If an error occurred, stop reading data.
double bp = partStrand->GetPairProbability( j, pair );
if( ( error = partChecker->isErrorStatus() ) ) { break; }
// Set the proper values for the color code.
probabilityAnnotations[i-1][j-1] =
( bp >= 0.99 ) ? 'a' :
( bp > 0.95 ) ? 'b' :
( bp > 0.90 ) ? 'c' :
( bp > 0.80 ) ? 'd' :
( bp > 0.70 ) ? 'e' :
( bp > 0.60 ) ? 'f' :
( bp > 0.50 ) ? 'g' :
'h';
probabilityAnnotations[i-1][pair-1] =
probabilityAnnotations[i-1][j-1];
}
}
}
}
// If an error occurred, print out an extra error message to make sure the
// user knows the error came from reading the partition function annotation
// file in.
if( error ) {
cerr << "Partition function save file not read successfully." << endl;
}
// Delete the RNA strand and error checker when they're no longer needed.
delete partStrand;
delete partChecker;
}
///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void Multilign_Interface::run() {
/*
* Rearrange the files vectors so they're organized by sequence number, not by file type.
* This is done behind the scenes; no progress message is necessary here.
*/
vector< vector<string> > matrix;
vector<string> filesSeq = files[0];
vector<string> filesCT = files[1];
vector<string> filesConstraint = files[2];
vector<string> filesSHAPE = files[3];
for( int i = 1; i <= seqNumber; i++ ) {
vector<string> row;
row.push_back( filesSeq[i-1] );
row.push_back( filesCT[i-1] );
row.push_back( filesConstraint[i-1] );
row.push_back( filesSHAPE[i-1] );
matrix.push_back( row );
}
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for Multilign_object that specifies a matrix of file names and a nucleic acid type.
* This allows for many varied sequence files to be used as input.
*
* After construction of the data structure, create the error checker which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant of the isErrorStatus method, which returns 0 if no error occurred.
* The calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
Multilign_object* object = new Multilign_object( matrix, isRNA );
ErrorChecker<Multilign_object>* checker = new ErrorChecker<Multilign_object>( object );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Set the temperature using the SetTemperature method.
* Only set the temperature if a given temperature doesn't equal the default.
* If the temperature does need to be set, use the error checker's isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && ( temperature != 310.15 ) ) {
// Show a message saying that the temperature is being set.
cout << "Setting temperature..." << flush;
// Set the temperature and check for errors.
object->SetTemperature( temperature );
error = checker->isErrorStatus();
// If no error occurred, print a message saying that temperature is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Randomize the sequences using the Randomize method.
*/
if( ( error == 0 ) && ( random == true ) ) {
// Show a message saying that randomization has started.
cout << "Randomizing sequences..." << flush;
// Randomize sequences.
// No error can be thrown from this method.
object->Randomize();
// Print a message saying that randomization is done.
cout << "done." << endl;
}
/*
* Set the index sequence using the SetIndexSeq method.
* Only set the index sequence if the given index sequence is not 1.
* After setting the index sequence, use the error checker's isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && indexSeq != 1 ) {
// Show a message saying that index sequence setting has started.
cout << "Setting index sequence..." << flush;
// Set the index sequence and check for errors.
int seqError = object->SetIndexSeq( indexSeq );
error = checker->isErrorStatus( seqError );
// If no error occurred, print a message saying that sequence setting is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Set the iterations using the SetIterations method.
*/
if( ( error == 0 ) && ( iterations != 2 ) ) {
// Show a message saying that iteration setting has started.
cout << "Setting iterations..." << flush;
// Set iterations.
int itError = object->SetIterations( iterations );
error = checker->isErrorStatus( itError );
// If no error occurred, print a message saying that iteration setting
// is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Set a maxdsvchange using the SetMaxDsv method.
*/
if( ( error == 0 ) && ( maxDsvChange != 1 ) ) {
// Show a message saying that max DSV change setting has started.
cout << "Setting max DSV change..." << flush;
// Set max DSV change.
int dsvError = object->SetMaxDsv( maxDsvChange );
error = checker->isErrorStatus( dsvError );
// If no error occurred, print a message saying that max dsv change setting is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Set max pairs using the SetMaxPairs method.
*/
if( ( error == 0 ) && ( maxPairs != -1 ) ) {
// Show a message saying that max pairs setting has started.
cout << "Setting max pairs..." << flush;
// Set max pairs.
int pairsError = object->SetMaxPairs( maxPairs );
error = checker->isErrorStatus( pairsError );
// If no error occurred, print a message saying that max pairs setting is done.
if( error == 0 ) { cout << "done." << endl; }
}
/**
* Set SHAPE data using the SetSHAPEIntercept and SetSHAPESlope method.
* Only do this if SHAPE data has been read.
*/
if( ( error == 0 ) && ( hasSHAPE == true ) ) {
// Show a message saying that SHAPE parameter setting has started.
cout << "Setting SHAPE parameters..." << flush;
// Set SHAPE parameters.
object->SetSHAPEIntercept( intercept );
object->SetSHAPESlope( slope );
// If no error occurred, print a message saying that SHAPE parameter setting is done.
cout << "done." << endl;
}
/*
* Run the Multilign algorithm using the ProgressiveMultilign method.
* During calculation, monitor progress using the TProgressDialog class and the Start/StopProgress methods of the RNA class.
* Neither of these methods require any error checking.
* After the main calculation is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the main calculation has started.
cout << "Folding nucleic acids..." << endl;
// Create the progress monitor.
TProgressDialog* progress = new TProgressDialog();
object->SetProgress( progress );
// Run the Multilign algorithm and check for errors.
int mainCalcError = object->ProgressiveMultilign(
processors, true, true, maxStructures, basepairWindow,
alignmentWindow, percent, maxSeparation, gap, inserts,
percentSingle, local );
error = checker->isErrorStatus( mainCalcError );
// Delete the progress monitor.
object->StopProgress();
delete progress;
// If no error occurred, print a message saying that the main calculation is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Write the alignment file using the WriteAlignment method.
* After writing is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the alignment file is being written.
cout << "Writing multiple alignment file..." << flush;
// Write the alignment file and check for errors.
int writeError = object->WriteAlignment( alignmentFile );
error = checker->isErrorStatus( writeError );
// If no errors occurred, print a message saying that alignment file writing is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* If requested by the user, clean up the intermediate files created by Multilign using the CleanupIntermediateFiles method.
* After files are cleaned up, use the error checker's isErrorStatus method to check for errors.
*/
if( keepIntermediate == false ) {
int cleanupError = object->CleanupIntermediateFiles();
error = checker->isErrorStatus( cleanupError );
}
// Delete the error checker and data structure.
delete checker;
delete object;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void DrawStructure::run() {
// Create a variable that tracks the result of calculations.
// Throughout, the calculation continues if result = "".
string result = "";
// Create variables to determine the range of structures that will be drawn.
// Initialize them so the range of structures will only be the specific structure given, if any.
int start = number;
int end = number;
// Create a temporary strand for the purpose of determining the number and range of structures to be drawn.
int structures = 0;
if( result == "" ) {
// Show a message that the information gathering phase has begun.
cout << "Determining structure information..." << flush;
// Create the temporary strand to get the number of structures, then delete it.
RNA* strand = new RNA( inputFile.c_str(), 1 );
ErrorChecker<RNA>* checker = new ErrorChecker<RNA>( strand );
result = checker->returnError();
structures = strand->GetStructureNumber();
delete strand;
delete checker;
// Use the number of structures to determine if the user gave an appropriate specific structure value, if necessary.
// If the structure value is invalid, set the error string to show this.
if( result == "" ) {
if( number != -1 && !( number >= 1 && number <= structures ) ) {
result = "Structure number out of range.\n";
}
}
// If the user did not give a specific structure, set the range of structures to all possible structures.
if( number == -1 ) {
start = 1;
end = structures;
}
// If no error occurred, print a message saying that the information gathering phase is done.
// Otherwise, print the error message.
if( result == "" ) {
cout << "done." << endl;
}
else {
cerr << endl << result;
}
}
// If no errors have occurred in structure preparation, draw the structure(s).
if( result == "" ) {
// Print a message saying that the drawing phase has started.
cout << "Drawing..." << endl;
// If the output file already exists, delete the file so a new file with that name can be generated.
ifstream test( outputFile.c_str() );
bool exists = test.good();
test.close();
if( exists ) {
remove( outputFile.c_str() );
}
// For each structure in the range, get its data and draw it appropriately.
for( int i = start; i <= end; i++ ) {
// Print a message saying that a particular structure has started drawing.
cout << " Drawing structure " << i << "..." << flush;
// Create the drawing handler and set whether it should circle nucleotides.
StructureImageHandler* handler = new StructureImageHandler();
handler->setNucleotidesCircled( encircle );
// Read in the structure.
// Note that if a pseudoknotted structure is fed to the radial routine, the radial routine automatically calls the circular routine.
if( result == "" ) {
if( circular ) {
result = handler->readCircular( inputFile, i );
}
else if( flat ) {
result = handler->readLinear( inputFile, i );
}
else {
handler->readRadial( inputFile, i );
}
}
// Read in annotation, if necessary.
if( result == "" ) {
if( probabilityFile != "" ) {
result = handler->addAnnotationProbability( probabilityFile, textAnnotation );
} else if( SHAPEFile != "" ) {
result = handler->addAnnotationSHAPE( SHAPEFile );
}
}
// Flip the structure, if asked.
if( ( result == "" ) && ( levorotatory == true ) ) {
handler->flipHorizontally();
}
// Write the structure image, as either SVG or Postscript.
if( result == "" ) {
if( isSVG ) {
handler->writeSVG( outputFile );
}
else {
bool append = ( number == -1 );
if( i == start ) {
ifstream test( outputFile.c_str() );
if( test ) {
append = false;
}
}
handler->writePostscript( outputFile, append );
}
}
// Delete the drawing handler.
delete handler;
// Print a message saying that a particular structure has finished drawing, if no error occurred.
// If an error did occur, break out of the loop.
if( result == "" ) {
cout << "done." << endl;
}
else break;
}
// If no error occurred, print a message saying that the drawing phase is done.
// Otherwise, print the error message.
if( result == "" ) {
cout << "done." << endl;
}
else {
cerr << endl << result << endl;
}
}
// Print confirmation of run finishing.
if( result == "" ) {
cout << calcType << " complete." << endl;
}
else {
cerr << calcType << " complete with errors." << endl;
}
}
///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void TurboFold_Interface::run() {
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for TurboFold_object that specifies vectors of file names.
* This allows for many varied sequence files to be used as input.
*
* After construction of the data structure, create the error checker which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant of the isErrorStatus method, which returns 0 if no error occurred.
* The calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
TurboFold* object = new TurboFold( &files[0], &files[2] );
ErrorChecker<TurboFold>* checker = new ErrorChecker<TurboFold>( object );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Set the temperature using the SetTemperature method.
* Only set the temperature if a given temperature doesn't equal the default.
* If the temperature does need to be set, use the error checker's isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && ( temperature != 310.15 ) ) {
// Show a message saying that the temperature is being set.
cout << "Setting temperature..." << flush;
// Set the temperature and check for errors.
int tempError = object->SetTemperature( temperature );
error = checker->isErrorStatus( tempError );
// If no error occurred, print a message saying temperature is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Set the maximum pairing distance, if required.
*/
if( ( error == 0 ) && ( distance > 0 ) ) {
// Show a message saying that maximum pairing distance is being set.
cout << "Setting maximum pairing distance..." << flush;
// Set the maximum pairing distance and check for errors.
int distError = object->SetMaxPairingDistance( distance );
error = checker->isErrorStatus( distError );
// If no error occurred, print a message saying that maximum pairing
// distance is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Read SHAPE data where necessary for specific sequence files.
* Since SHAPE data must be read individually for each sequence, use the ReadSHAPE method.
*/
if( error == 0 && hasSHAPE ) {
// Show a message saying SHAPE data is being read.
cout << "Reading SHAPE data..." << flush;
// For each sequence, read in SHAPE data, if applicable.
for( int i = 1; i <= seqNumber; i++ ) {
if( files[3][i-1] != "" ) {
int shapeError = object->ReadSHAPE( i, files[3][i-1].c_str(), slope, intercept );
if( ( error = checker->isErrorStatus( shapeError ) ) == true ) {
i += seqNumber;
}
}
}
// Show a message saying SHAPE data reading is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Run the TurboFold algorithm using the fold method.
* During calculation, monitor progress using the TProgressDialog class and the Start/StopProgress methods of the RNA class.
* Neither of these methods require any error checking.
* After the main calculation is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the main calculation has started.
cout << "Folding nucleic acids..." << endl;
// Create the progress monitor.
TProgressDialog* progress = new TProgressDialog();
object->SetProgress( *progress );
// Run the TurboFold algorithm and check for errors.
int mainCalcError = object->fold( turboGamma, turboIterations, processors );
error = checker->isErrorStatus( mainCalcError );
// Delete the progress monitor.
object->StopProgress();
delete progress;
// If no error occurred, print message that main calculation is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Resolve the structures generated by TurboFold using specific methods, depending on the mode selected for TurboFold.
* In MEA mode, use the MaximizeExpectedAccuracy method.
* In ProbKnot mode, use the ProbKnot method.
* In Threshold mode, use the PredictProbablePairs method.
* During calculation, monitor progress using the TProgressDialog class and the Start/StopProgress methods of the RNA class.
* Neither of these methods require any error checking.
* After the resolution calculation is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the resolving calculation has started.
if( mode == "MEA" ) {
cout << "Calculating maximum expected accuracy structures..." << flush;
} else if( mode == "ProbKnot" ) {
cout << "Predicting pseudoknots..." << flush;
} else {
cout << "Calculating probable pairs..." << flush;
}
// Resolve the structures and check for errors.
for( int i = 1; i <= seqNumber; i++ ) {
int resolveError = 0;
if( mode == "MEA" ) {
resolveError = object->MaximizeExpectedAccuracy( i, percent, maxStructures, windowSize, meaGamma );
} else if( mode == "ProbKnot" ) {
resolveError = object->ProbKnot( i, pkIterations, minHelixLength );
} else {
resolveError = object->PredictProbablePairs( i, threshold );
}
error = checker->isErrorStatus( resolveError );
if( error != 0 ) { i += seqNumber; }
}
// If no error occurred, print message that resolving is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Write CT output files using the WriteCt method.
* After writing is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the CT files are being written.
cout << "Writing output ct files..." << flush;
// Write the CT files and check for errors.
for( int i = 1; i <= seqNumber; i++ ) {
int writeError = object->WriteCt( i, files[1][i-1].c_str() );
error = checker->isErrorStatus( writeError );
if( error != 0 ) { i += seqNumber; }
}
// If no errors occurred, show a CT files writing completion message.
if( error == 0 ) { cout << "done." << endl; }
}
// Delete the error checker and data structure.
delete checker;
delete object;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
///////////////////////////////////////////////////////////////////////////////
// Run bimolecular folding calculations.
///////////////////////////////////////////////////////////////////////////////
void bifold::run() {
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for HybridRNA that specifies two filenames and types.
* For both sequences, specify type = 2 (sequence file).
* isRNA identifies whether the strand is RNA (true) or DNA (false).
*
* After construction of the strand data structure, create the error checker
* which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant
* of the isErrorStatus method, which returns 0 if no error occurred. The
* calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
HybridRNA* strand =
new HybridRNA( seqFile1.c_str(), 2, seqFile2.c_str(), 2, isRNA );
ErrorChecker<HybridRNA>* checker = new ErrorChecker<HybridRNA>( strand );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Set the temperature using the SetTemperature method.
* Only set the temperature if a given temperature doesn't equal the default.
* If the temperature does need to be set, use the error checker's
* isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && ( temperature != 310.15 ) ) {
// Show a message saying that the temperature is being set.
cout << "Setting temperature..." << flush;
// Set the temperature and check for errors.
int tempError = strand->SetTemperature( temperature );
error = checker->isErrorStatus( tempError );
// If no error occurred, print a message saying that temperature is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Set allowance or denial of intramolecular pairs using the
* SetForbidIntramolecular method.
* Since this method only sets a flag, error checking is not necessary.
*/
if( error == 0 ) {
strand->SetForbidIntramolecular( intramolecular );
}
/*
* Fold the hybrid strand using the FoldBimolecular method.
* If a save file name has been specified, the FoldBimolecular method also
* has the ability to write a folding save file.
* During calculation, monitor progress using the TProgressDialog class and
* the Start/StopProgress methods of the RNA class. Neither of these methods
* require any error checking.
* After the main calculation is complete, use the error checker's
* isErrorStatus method to check for errors in the main calculation.
*/
if( error == 0 ) {
// Show a message saying that the main calculation has started.
cout << "Folding two strands..." << endl;
// Create the progress monitor.
TProgressDialog* progress = new TProgressDialog();
strand->SetProgress( *progress );
// Fold the hybrid strand and check for errors.
int mainCalcError =
strand->FoldBimolecular( percent, maxStructures, windowSize,
saveFile.c_str(), maxLoop );
error = checker->isErrorStatus( mainCalcError );
// Delete the progress monitor.
strand->StopProgress();
delete progress;
// If no error occurred, print message that main calculation is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Write a CT output file using the WriteCt method.
* After writing is complete, use the error checker's isErrorStatus method to
* check for errors.
*/
if( error == 0 ) {
// Show a message saying that the CT file is being written.
cout << "Writing output ct file..." << flush;
// Write the CT file and check for errors.
int writeError = strand->WriteCt( ctFile.c_str() );
error = checker->isErrorStatus( writeError );
// If no errors occurred, show a CT file writing completion message.
if( error == 0 ) { cout << "done." << endl; }
}
// Delete the error checker and data structure.
delete checker;
delete strand;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void refold::run() {
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for RNA that specifies a filename.
* Specify type = 4 (folding save file).
* The save file handles the type of nucleic acid being folded, so it can be set as the default (RNA) in the constructor.
*
* After construction of the strand data structure, create the error checker which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant of the isErrorStatus method, which returns 0 if no error occurred.
* The calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
RNA* strand = new RNA( saveFile.c_str(), 4 );
ErrorChecker<RNA>* checker = new ErrorChecker<RNA>( strand );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Set the window size, based on the length of the sequence given as input.
* Only do this if window size hasn't been set on the command line.
* Use method GetSequenceLength to get the length of the sequence.
* The window sizes in relation to the length are hardcoded values.
*/
if( windowSize == -1 && error == 0 ) {
int length = strand->GetSequenceLength();
windowSize =
( length > 1200 ) ? 20 :
( length > 800 ) ? 15 :
( length > 500 ) ? 11 :
( length > 300 ) ? 7 :
( length > 120 ) ? 5 :
( length > 50 ) ? 3 :
2;
}
/*
* Refold the regenerated strand using the ReFoldSingleStrand method.
* After the main calculation is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the main calculation has started.
cout << "Refolding nucleic acids..." << flush;
// Do the main calculation and check for errors.
int mainCalcError = strand->ReFoldSingleStrand( percent, maxStructures, windowSize );
error = checker->isErrorStatus( mainCalcError );
// If no error occurred, print a message saying that the main calculation is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Write a CT output file using the WriteCt method.
* After writing is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the CT file is being written.
cout << "Writing output ct file..." << flush;
// Write the CT file and check for errors.
int writeError = strand->WriteCt( ctFile.c_str() );
error = checker->isErrorStatus( writeError );
// If no errors occurred, show a CT file writing completion message.
if( error == 0 ) { cout << "done." << endl; }
}
// Delete the error checker and data structure.
delete checker;
delete strand;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
///////////////////////////////////////////////////////////////////////////////
// Run calculations.
///////////////////////////////////////////////////////////////////////////////
void DuplexFold::run() {
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for HybridRNA that specifies two filenames and types.
* For both sequences, specify type = 2 (sequence file).
* isRNA identifies whether the strand is RNA (true) or DNA (false).
*
* After construction of the strand data structure, create the error checker which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant of the isErrorStatus method, which returns 0 if no error occurred.
* The calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
HybridRNA* strand = new HybridRNA( seqFile1.c_str(), 2, seqFile2.c_str(), 2, isRNA );
ErrorChecker<HybridRNA>* checker = new ErrorChecker<HybridRNA>( strand );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Set the temperature using the SetTemperature method.
* Only set the temperature if a given temperature doesn't equal the default.
* If the temperature does need to be set, use the error checker's isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && ( temperature != 310.15 ) ) {
// Show a message saying that the temperature is being set.
cout << "Setting temperature..." << flush;
// Set the temperature and check for errors.
int tempError = strand->SetTemperature( temperature );
error = checker->isErrorStatus( tempError );
// If no error occurred, print a message saying that temperature is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Fold the hybrid strand using the FoldDuplex method.
* After the main calculation is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the main calculation has started.
cout << "Folding duplex..." << endl;
// Create the progress monitor.
TProgressDialog* progress = new TProgressDialog();
strand->SetProgress( *progress );
// Do the main calculation and check for errors.
int mainCalcError = strand->FoldDuplex( percent, maxStructures, windowSize, maxLoop );
error = checker->isErrorStatus( mainCalcError );
// Delete the progress monitor.
strand->StopProgress();
delete progress;
// If no error occurred, print message that main calculation is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Write a CT output file using the WriteCt method.
* After writing is complete, use the error checker's isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the CT file is being written.
cout << "Writing output ct file..." << flush;
// Write the CT file and check for errors.
int writeError = strand->WriteCt( ctFile.c_str() );
error = checker->isErrorStatus( writeError );
// If no errors occurred, show a CT file writing completion message.
if( error == 0 ) { cout << "done." << endl; }
}
// Delete the error checker and data structure.
delete checker;
delete strand;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
///////////////////////////////////////////////////////////////////////////////
// Run the suboptimal structure calculations.
///////////////////////////////////////////////////////////////////////////////
void AllSub::run() {
// Create a variable that handles errors.
int error = 0;
/*
* Use the constructor for RNA that specifies a filename.
* Specify type = 2 (sequence file).
* isRNA identifies whether the strand is RNA (true) or DNA (false).
*
* After construction of the strand data structure, create the error checker
* which monitors for errors.
* Throughout, the error status of the calculation is checked with a variant
* of the isErrorStatus method, which returns 0 if no error occurred. The
* calculation proceeds as long as error = 0.
*/
cout << "Initializing nucleic acids..." << flush;
RNA* strand = new RNA( seqFile.c_str(), 2, isRNA );
ErrorChecker<RNA>* checker = new ErrorChecker<RNA>( strand );
error = checker->isErrorStatus();
if( error == 0 ) { cout << "done." << endl; }
/*
* Set the percent difference and the maximum absolute energy difference.
* Both differences are based on the length of the sequence, and their values
* in relation to the sequence length are hardcoded.
* Only set these values if they were not set on the command line. As one or
* the other of them may or may not be set, each one is checked individually.
*
* Get the length of the sequence using the GetSequenceLength method.
* Since this method only returns a length, error checking is not necessary.
*/
if( error == 0 ) {
// Get the sequence length to identify the hardcoded thresholds.
int length = strand->GetSequenceLength();
// Set the maximum percent difference, if applicable.
if( percent == -1 ) {
percent =
( length > 1200 ) ? 5 :
( length > 800 ) ? 8 :
( length > 500 ) ? 10 :
( length > 300 ) ? 15 :
( length > 120 ) ? 20 :
( length > 50 ) ? 25 :
50;
}
// Set the absolute energy difference, if applicable.
if( absolute == -1 ) {
absolute =
( length > 1200 ) ? 0.25 :
( length > 800 ) ? 0.5 :
( length > 500 ) ? 0.75 :
( length > 300 ) ? 1 :
( length > 120 ) ? 1.5 :
( length > 50 ) ? 3 :
10;
}
}
/*
* Set the temperature using the SetTemperature method.
* Only set the temperature if a given temperature doesn't equal the default.
* If the temperature does need to be set, use the error checker's
* isErrorStatus method to check for errors.
*/
if( ( error == 0 ) && ( temperature != 310.15 ) ) {
// Show a message saying that the temperature is being set.
cout << "Setting temperature..." << flush;
// Set the temperature and check for errors.
int tempError = strand->SetTemperature( temperature );
error = checker->isErrorStatus( tempError );
// If no error occurred, print a message saying that temperature is set.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Add constraints if a file has been given for their inclusion.
* Read in this constraints with method ReadConstraints.
* After constraints are read, use the error checker's isErrorStatus method
* to check for errors.
*/
if( error == 0 && constraintFile != "" ) {
// Show a message saying that constraints are being applied.
cout << "Applying constraints..." << flush;
// Apply constraints and check for errors.
int constraintError = strand->ReadConstraints( constraintFile.c_str() );
error = checker->isErrorStatus( constraintError );
// If no error occurred, print a message saying constraints were included.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Generate structures using the GenerateAllSuboptimalStructures method.
* During calculation, monitor progress using the TProgressDialog class and
* the Start/StopProgress methods of the RNA class. Neither of these methods
* require any error checking.
* After the main calculation is complete, use the error checker's
* isErrorStatus method to check for errors.
*/
if( error == 0 ) {
// Show a message saying that the main calculation has started.
cout << "Generating suboptimal structures..." << endl;
// Create the progress monitor.
TProgressDialog* progress = new TProgressDialog();
strand->SetProgress( *progress );
// Generate the suboptimal structures and check for errors.
int mainCalcError =
strand->GenerateAllSuboptimalStructures( (float)percent, absolute );
error = checker->isErrorStatus( mainCalcError );
// Delete the progress monitor.
strand->StopProgress();
delete progress;
// If no error occurred, print message that main calculation is done.
if( error == 0 ) { cout << "done." << endl; }
}
/*
* Write a CT output file using the WriteCt method.
* After writing is complete, use the error checker's isErrorStatus method to
* check for errors.
*/
if( error == 0 ) {
// Show a message saying that the CT file is being written.
cout << "Writing output ct file..." << flush;
// Write the CT file and check for errors.
int writeError = strand->WriteCt( ctFile.c_str() );
error = checker->isErrorStatus( writeError );
// If no errors occurred, show a CT file writing completion message.
if( error == 0 ) { cout << "done." << endl; }
}
// Delete the error checker and data structure.
delete checker;
delete strand;
// Print confirmation of run finishing.
if( error == 0 ) { cout << calcType << " complete." << endl; }
else { cerr << calcType << " complete with errors." << endl; }
}
//fetch each object, error checking, build row
//input: raw string of row from CSV, column index for data type
//output: complete row object
Grant_rowObject GrantRowBuilder::buildRow(string data, ColIndex index){
//*** instantiate classes that will help out in this function
AttributeRetriever fetch(data);
bool hasError = 0;//IF this row has an error in it, this flag should be set to true
ErrorChecker filter;
bool enableErrorChecking (1);
//*** retrieve each attribute - temporarily store them
string name = fetch.getAttribute(index.name_loc);
string domain = fetch.getAttribute(index.domain_loc);
string fundType = fetch.getAttribute(index.fundType_loc);
string stat = fetch.getAttribute(index.stat_loc);
string role = fetch.getAttribute(index.role_loc);
string title = fetch.getAttribute(index.title_loc);
string pInvestigator = fetch.getAttribute(index.pInvestigator_loc);
// Deal with multiple co-investigators in a cell
string cpInvestigator = fetch.getAttribute(index.cpInvestigator_loc);
fetch.grabFirstString(cpInvestigator);//TEMPORARY -returns first co-investigator
bool peerReviewed = fetch.getBoolAttribute(index.peerReviewed_loc);
bool indGrant = fetch.getBoolAttribute(index.indGrant_loc);
int sDate = fetch.getIntAttribute(index.sDate_loc);
int eDate = fetch.getIntAttribute(index.eDate_loc);
long long totalAmount = fetch.getLongAttribute(index.totalAmount_loc);
//*** Error Checking
if (enableErrorChecking){
//strings - chack for blank fields
name = filter.blankCatch(name, hasError);
domain = filter.blankCatch(domain, hasError);
fundType = filter.blankCatch(fundType, hasError);
stat = filter.blankCatch(stat, hasError);
role = filter.blankCatch(role, hasError);
title = filter.blankCatch(title, hasError);
pInvestigator = filter.blankCatch(pInvestigator, hasError);
//PEER REVIEWED - if feild is left blank, we assume it has not been peer reviewed
//INDUSTRY GRANT - if feild is left blank, we assume it was not an industry grant
//dates - check for zeroes
sDate = filter.zeroCatch(sDate, hasError);
eDate = filter.zeroCatch(eDate, hasError);
//Assert FundType
string acceptable_fundTypes [2] = {"Grants","Clinical Trials"};
fundType = filter.stringAssert(fundType, hasError, 2, acceptable_fundTypes);
//Assert status
string acceptable_statuses [5] = {"Applied","Funded","Declined","Rejected","Not Funded"};
stat = filter.stringAssert(stat, hasError, 5, acceptable_statuses);
//Assert Role
string acceptable_Roles [8] = {"Co-Applicant","Co-Investigator","Collaborator","Co-Principal Investigator","Principal Applicant","Principal Investigator","Principal Site Investigator","Site Investigator"};
role = filter.stringAssert(role,hasError, 8, acceptable_Roles);
}
//*** Build Row
Grant_rowObject currentRow (hasError, name, domain, sDate, eDate, fundType, stat, peerReviewed,indGrant, role, title, pInvestigator, cpInvestigator, totalAmount);
return currentRow;
};
