Expression* Expression :: Compile (const RGString& expressionString) {
int errcnt = 0;
finalExpression = NULL;
RemainingCharacters = (int)expressionString.Length ();
ExpressionCriterion = expressionString;
name_stack.ClearAndDelete ();
if ( Exprparse ())
errcnt++;
if (errcnt > 0)
cout << "Found errors in: " << expressionString.GetData () << endl;
if (name_stack.Entries () > 0)
cout << "Left with " << name_stack.Entries () << " names still in stack...compilation incomplete." << endl;
name_stack.ClearAndDelete ();
if (errcnt == 0)
return finalExpression;
return NULL;
}
RGPersistent* RGHashTable :: RemoveAllReferences (const RGPersistent* p) {
RGDList* list = SelectList (p);
int NEntriesStart = list->Entries ();
int NEntriesEnd;
RGPersistent* q = list->RemoveAllReferences (p);
NEntriesEnd = list->Entries ();
NumberOfEntries -= (NEntriesStart - NEntriesEnd);
return q;
}
Boolean RGXmlDataMinLength :: Test (const RGDList& x) const {
if (x.Entries () >= SizeLimit)
return TRUE;
return FALSE;
}
bool GenotypeSet :: FindAllGenotypes (const RGString& fileName, RGDList& genotypes) {
RGDListIterator it (mGenotypes);
GenotypesForAMarkerSet* nextSet;
IndividualGenotype* ig;
int startEntries = genotypes.Entries ();
int endEntries;
while (nextSet = (GenotypesForAMarkerSet*) it ()) {
ig = nextSet->FindGenotypeForFileName (fileName);
if (ig != NULL)
genotypes.Append (ig);
}
endEntries = genotypes.Entries ();
if (endEntries > startEntries)
return true;
return false;
}
bool LocusCollection :: FindAllCopiesOfNamedLocus (const RGString& name, RGDList& results) {
RGDListIterator it (mLociByName);
IndividualLocus* nextLocus;
while (nextLocus = (IndividualLocus*) it ()) {
if (name == nextLocus->GetName ())
results.Append (nextLocus);
}
if (results.Entries () > 0)
return true;
return false;
}
bool LocusCollection :: FindAllAllelesForNamedLocus (const RGString& name, RGDList& results) {
RGDList temp;
if (!FindAllCopiesOfNamedLocus (name, temp))
return false;
IndividualLocus* nextLocus;
while (nextLocus = (IndividualLocus*) temp.GetFirst ())
nextLocus->AppendAllelesToList (results);
if (results.Entries () == 0)
return false;
return true;
}
STRBaseLaneStandard :: STRBaseLaneStandard (const RGString& xmlString) : BaseLaneStandard (xmlString),
mCharacteristicArray (NULL), mDifferenceArray (NULL), mUnnormalizedDifferenceArray(0),
mNormsLeft (NULL), mNormsRight (NULL), mRelativeSizes (NULL),
mSize (0), mDifferenceSize (0), mOmissionSize (0), mOmissionArray (NULL), mNumberOfLargePeaks (0),
mLargeCharacteristicArray (NULL), mLargeDifferenceArray (NULL), mNumberOfLargeDifferences (0),
mFirstIntervalFraction (-1.0), mSmallestIntervalFraction (-1.0), mMaxRelativeHeight (-1),
mCorrelationAcceptanceThreshold (0.993), mCorrelationAutoAcceptanceThreshold (0.9975)
//
//, mLastHalfSize (0), mLastHalfCharacteristicArray (NULL), mLastHalfUnnormalizedDifferenceArray (NULL),
//mLastHalfRelativeSizes (NULL), mLastHalfDifferenceSize (0), mLastHalfNormsLeft (NULL), mLastHalfNormsRight (NULL)
{
RGString TextInput (xmlString);
RGBracketStringSearch CharacteristicsToken ("<Characteristics>", "</Characteristics>", TextInput);
RGString CharacteristicsString;
RGString HeightString;
RGBracketStringSearch HeightToken ("<RelativeHeights>", "</RelativeHeights>", TextInput);
RGBracketStringSearch OmissionSearch ("<Omissions>", "</Omissions>", TextInput);
RGBracketStringSearch AcceptanceThresholdSearch ("<CorrelationAcceptanceThreshold>", "</CorrelationAcceptanceThreshold>", TextInput);
RGBracketStringSearch AutoAcceptanceThresholdSearch ("<CorrelationAutoAcceptanceThreshold>", "</CorrelationAutoAcceptanceThreshold>", TextInput);
RGString OmissionString;
size_t endPosition;
RGString token;
RGString delim;
RGString numberString;
double minChar;
double maxChar;
RGDList tokens;
double value;
RGPDouble* valuePtr;
double previous;
double next;
double cumulative;
int i = 0;
double temp;
int k;
int j;
if (!CharacteristicsToken.FindNextBracketedString (0, endPosition, CharacteristicsString)) {
Msg << "Internal Lane Standard named " << Name << " could not find Characteristics token from string\n";
Msg << TextInput << "\n";
Valid = FALSE;
}
else {
RGStringTokenizer data (CharacteristicsString);
data.AddDelimiter (' ');
data.AddDelimiter (" ");
data.AddDelimiter (" ");
data.AddDelimiter (" ");
data.AddDelimiter (" ");
data.AddRemoveItem (',');
data.AddDelimiter ('\t');
data.AddDelimiter ('\n');
while (data.NextToken (token, delim)) {
if (token.Length () > 0) {
value = token.ConvertToDouble ();
valuePtr = new RGPDouble (value);
if (tokens.Entries () == 0) {
minChar = maxChar = value;
tokens.Append (valuePtr);
}
else if (value < minChar) {
minChar = value;
tokens.Prepend (valuePtr);
}
else if (value > maxChar) {
maxChar = value;
tokens.Append (valuePtr);
}
else
tokens.Insert (valuePtr);
}
}
mSize = (int)tokens.Entries ();
NCharacteristics = mSize;
mCharacteristicArray = new double [mSize];
mDifferenceSize = mSize - 1;
mDifferenceArray = new double [mDifferenceSize];
mUnnormalizedDifferenceArray = new double [mDifferenceSize];
mNormsLeft = new double [mDifferenceSize];
mNormsRight = new double [mDifferenceSize];
//mLastHalfSize = mSize / 2;
//if (2 * mLastHalfSize < mSize)
// mLastHalfSize += 1;
//mLastHalfDifferenceSize = mLastHalfSize - 1;
//mLastHalfCharacteristicArray = new double [mLastHalfSize];
//mLastHalfUnnormalizedDifferenceArray = new double [mLastHalfDifferenceSize];
//mLastHalfNormsLeft = new double [mLastHalfDifferenceSize];
//mLastHalfNormsRight = new double [mLastHalfDifferenceSize];
while (valuePtr = (RGPDouble*)tokens.GetFirst ()) {
mCharacteristicArray [i] = valuePtr->GetDouble ();
delete valuePtr;
i++;
}
//int displacement = mSize - mLastHalfSize; // +0, +1, -1?????
//for (i=0; i<mLastHalfSize; i++)
// mLastHalfCharacteristicArray [i] = mCharacteristicArray [i + displacement];
previous = mCharacteristicArray [0];
for (i=0; i<mDifferenceSize; i++) {
next = mCharacteristicArray [i+1];
mUnnormalizedDifferenceArray [i] = mDifferenceArray [i] = next - previous;
previous = next;
}
//for (i=0; i<mLastHalfSize; i++)
// mLastHalfUnnormalizedDifferenceArray [i] = mUnnormalizedDifferenceArray [i + displacement];
double norm = 0.0;
for (i=0; i<mDifferenceSize; i++)
norm += mDifferenceArray [i] * mDifferenceArray [i];
norm = sqrt (norm);
if (norm > 0.0) {
for (i=0; i<mDifferenceSize; i++)
mDifferenceArray [i] /= norm;
}
cumulative = 0.0;
for (i=0; i<mDifferenceSize; i++) {
temp = mUnnormalizedDifferenceArray [i];
cumulative += temp * temp;
mNormsLeft [i] = cumulative;
}
cumulative = 0.0;
for (i=mDifferenceSize-1; i>=0; i--) {
temp = mUnnormalizedDifferenceArray [i];
cumulative += temp * temp;
mNormsRight [i] = cumulative;
}
//cumulative = 0.0;
//for (i=0; i<mLastHalfDifferenceSize; i++) {
// temp = mLastHalfUnnormalizedDifferenceArray [i];
// cumulative += temp * temp;
// mLastHalfNormsLeft [i] = cumulative;
//}
//cumulative = 0.0;
//for (i=mLastHalfDifferenceSize-1; i>=0; i--) {
// temp = mLastHalfUnnormalizedDifferenceArray [i];
// cumulative += temp * temp;
// mLastHalfNormsRight [i] = cumulative;
//}
MaxCharacteristic = maxChar;
MinCharacteristic = minChar;
double width = mCharacteristicArray [mSize - 1] - mCharacteristicArray [0];
if (width > 0.0)
mFirstIntervalFraction = mUnnormalizedDifferenceArray [0] / width;
else
mFirstIntervalFraction = 1.0;
double temp = mUnnormalizedDifferenceArray [0];
double value;
for (i=0; i<mDifferenceSize; i++) {
value = mUnnormalizedDifferenceArray [i];
if (value < temp)
temp = value;
}
if (width > 0.0)
mSmallestIntervalFraction = temp / width;
else
mSmallestIntervalFraction = 1.0;
if (!HeightToken.FindNextBracketedString (0, endPosition, HeightString)) {
mRelativeSizes = NULL;
// mLastHalfRelativeSizes = NULL;
}
else {
RGStringTokenizer data2 (HeightString);
data2.AddDelimiter (' ');
data2.AddDelimiter (" ");
data2.AddDelimiter (" ");
data2.AddDelimiter (" ");
data2.AddDelimiter (" ");
data2.AddRemoveItem (',');
data2.AddDelimiter ('\t');
data2.AddDelimiter ('\n');
mRelativeSizes = new int [mSize];
for (k=0; k<mSize; k++)
mRelativeSizes [k] = -1;
k = 0;
while (data2.NextToken (token, delim)) {
if (token.Length () > 0) {
token.ToUpper ();
char first = token.GetFirstCharacter ();
switch (first) {
case 'H':
mRelativeSizes [k] = 5;
break;
case 'L':
mRelativeSizes [k] = 1;
break;
case 'M':
if (token.Length () == 1)
mRelativeSizes [k] = 3;
else if (token.GetLastCharacter () == 'H')
mRelativeSizes [k] = 4;
else
mRelativeSizes [k] = 2;
break;
default:
mRelativeSizes [k] = -1;
}
}
k++;
if (k >= mSize)
break;
}
//mLastHalfRelativeSizes = new int [mLastHalfSize];
//for (i=0; i<mLastHalfSize; i++)
// mLastHalfRelativeSizes [i] = mRelativeSizes [i + displacement];
int maxRelativeSize = 0;
int relativeSize;
for (j=0; j<mSize; j++) {
relativeSize = mRelativeSizes [j];
if (relativeSize > maxRelativeSize)
maxRelativeSize = relativeSize;
}
int numberOfHighsInILS = 0;
mMaxRelativeHeight = maxRelativeSize;
for (j=0; j<mSize; j++) {
if (mRelativeSizes [j] == maxRelativeSize)
numberOfHighsInILS++;
}
mNumberOfLargePeaks = numberOfHighsInILS;
mNumberOfLargeDifferences = numberOfHighsInILS - 1;
mLargeCharacteristicArray = new double [numberOfHighsInILS];
if (mNumberOfLargeDifferences > 0)
mLargeDifferenceArray = new double [mNumberOfLargeDifferences];
k = 0;
for (j=0; j<mSize; j++) {
if (mRelativeSizes [j] == maxRelativeSize) {
mLargeCharacteristicArray [k] = mCharacteristicArray [j];
k++;
}
}
if (mLargeDifferenceArray != NULL) {
for (j=0; j<mNumberOfLargeDifferences; j++)
mLargeDifferenceArray [j] = mLargeCharacteristicArray [j + 1] - mLargeCharacteristicArray [j];
}
}
if (OmissionSearch.FindNextBracketedString (0, endPosition, OmissionString)) {
RGStringTokenizer omissionData (OmissionString);
omissionData.AddDelimiter (' ');
omissionData.AddDelimiter (" ");
omissionData.AddDelimiter (" ");
omissionData.AddDelimiter (" ");
omissionData.AddDelimiter (" ");
omissionData.AddRemoveItem (',');
omissionData.AddDelimiter ('\t');
omissionData.AddDelimiter ('\n');
while (omissionData.NextToken (token, delim)) {
if (token.Length () > 0) {
value = token.ConvertToDouble ();
valuePtr = new RGPDouble (value);
if (tokens.Entries () == 0) {
minChar = maxChar = value;
tokens.Append (valuePtr);
}
else if (value < minChar) {
minChar = value;
tokens.Prepend (valuePtr);
}
else if (value > maxChar) {
maxChar = value;
tokens.Append (valuePtr);
}
else
tokens.Insert (valuePtr);
}
}
mOmissionSize = (int)tokens.Entries ();
if (mOmissionSize == 0)
return;
mOmissionArray = new double [mOmissionSize];
i = 0;
while (valuePtr = (RGPDouble*)tokens.GetFirst ()) {
mOmissionArray [i] = valuePtr->GetDouble ();
delete valuePtr;
i++;
}
}
}
cout << "ILS Name = " << Name.GetData () << endl;
if (!AcceptanceThresholdSearch.FindNextBracketedString (0, endPosition, numberString))
cout << "\tCould not find ILS acceptance threshold...using " << mCorrelationAcceptanceThreshold << endl;
else {
mCorrelationAcceptanceThreshold = numberString.ConvertToDouble ();
cout << "\tFrom ILS settings...ILS acceptance threshold = " << mCorrelationAcceptanceThreshold << endl;
}
if (!AutoAcceptanceThresholdSearch.FindNextBracketedString (0, endPosition, numberString))
cout << "\tCould not find ILS auto acceptance threshold...using default value = " << mCorrelationAutoAcceptanceThreshold << endl << endl;
else {
mCorrelationAutoAcceptanceThreshold = numberString.ConvertToDouble ();
cout << "\tFrom ILS settings...ILS auto acceptance threshold = " << mCorrelationAutoAcceptanceThreshold << endl << endl;
}
}
Boolean STRBaseLocus :: ExtractGridSignals (RGDList& channelSignalList, const LaneStandard* ls, RGDList& locusDataSignalList, ChannelData* lsData, Locus* locus) {
int lsSize = ls->GetNumberOfCharacteristics ();
int i1;
int i2;
double c1;
double c2;
Notice* newNotice;
if (lsSize < 0) {
Msg << "Number of characteristics not available for Locus named " << LocusName << "\n";
newNotice = new SetupErrorNumberPeaksUnavailableForLocus;
locus->AddNoticeToList (newNotice);
return FALSE;
}
if (LowerBoundGridLSBasePair < 0.0) {
i1 = (int) floor (LowerBoundGridLSIndex);
if (i1 < 0) {
Msg << "Internal Lane Standard lower bound index for Grid is out of bounds for Locus named " << LocusName << "\n";
return FALSE;
}
i2 = i1 + 1;
c2 = LowerBoundGridLSIndex - i1;
c1 = 1.0 - c2;
MinimumGridTime = c1 * ls->GetLaneStandardTimeForCharacteristicNumber (i1) +
c2 * ls->GetLaneStandardTimeForCharacteristicNumber (i2);
// cout << "Not using base pairs for minimum of locus " << LocusName << endl;
}
else
MinimumGridTime = lsData->GetTimeForSpecifiedID (LowerBoundGridLSBasePair);
if (UpperBoundGridLSBasePair < 0.0) {
i1 = (int) floor (UpperBoundGridLSIndex);
if (i1 < 0) {
Msg << "Internal Lane Standard upper bound index for Grid is out of bounds for Locus named " << LocusName << "\n";
return FALSE;
}
i2 = i1 + 1;
c2 = UpperBoundGridLSIndex - i1;
c1 = 1.0 - c2;
MaximumGridTime = c1 * ls->GetLaneStandardTimeForCharacteristicNumber (i1) +
c2 * ls->GetLaneStandardTimeForCharacteristicNumber (i2);
// cout << "Not using base pairs for maximum of locus " << LocusName << endl;
}
else
MaximumGridTime = lsData->GetTimeForSpecifiedID (UpperBoundGridLSBasePair);
RGDListIterator it (channelSignalList);
DataSignal* nextSignal;
double nextMean;
locusDataSignalList.Clear ();
// Assume channelSignalList is in increasing order by signal means
while (nextSignal = (DataSignal*) it ()) {
nextMean = nextSignal->GetMean ();
if (nextMean >= MinimumGridTime) {
if (nextMean <= MaximumGridTime) {
it.RemoveCurrentItem ();
locusDataSignalList.Append (nextSignal);
}
else
break;
}
}
if (locusDataSignalList.Entries () < mLocusSize) {
Msg << "Locus named " << LocusName << " could not find the required number of unanalyzed peaks in range\n";
newNotice = new LocusHasTooFewPeaks;
locus->AddNoticeToList (newNotice);
return FALSE;
}
return TRUE;
}
int Ladder :: AmendLadderData (LadderInputFile* inFile, RGString& oldLadderString) {
RGString newLadderString;
// Parse oldLadder data into pieces for individual edits
RGString locusString;
RGString* newLocusString;
RGDList locusStrings;
size_t startPos = 0;
size_t endPos;
RGXMLTagSearch locusSearch ("Locus", oldLadderString);
RGXMLTagSearch searchRegionsSearch ("SearchRegions", oldLadderString);
RGString* ilsName = (RGString*)inFile->GetILSNameList ().First ();
endPos = 0;
oldLadderString.FindNextSubstring (0, "\t\t\t<Locus>", endPos);
RGString insertBase;
insertBase << "\t\t\t\t\t<ILSName>" << ilsName->GetData () << "</ILSName>\n";
insertBase << "\t\t\t\t</LSBases>";
RGString leadString = oldLadderString.ExtractSubstring (0, endPos - 1);
//cout << "Lead string = \n" << leadString.GetData () << endl;
endPos = 0;
leadString.FindAndReplaceNextSubstring ("\t\t\t\t</LSBases>", insertBase, endPos);
//cout << "Lead string = \n" << leadString.GetData () << endl;
newLadderString << leadString;
startPos = 0;
while (locusSearch.FindNextTag (startPos, endPos, locusString)) {
newLocusString = new RGString (locusString);
locusStrings.Append (newLocusString);
startPos = endPos;
}
if (mLocusList.size () != locusStrings.Entries ()) {
cout << "Number of loci in bins file does not match number of loci in ladder file" << endl;
return -152;
}
Locus* nextLocus;
RGString locusInsert;
RGString currentLocusString;
RGString nameString;
RGXMLTagSearch locusNameSearch ("Name", currentLocusString);
RGXMLTagSearch coreRepeatSearch ("CoreRepeatNumber", currentLocusString);
double minSearch;
double maxSearch;
RGString repeatString;
int repeatNumber;
while (locusStrings.Entries () > 0) {
newLocusString = (RGString*) locusStrings.GetFirst ();
currentLocusString = *newLocusString;
locusNameSearch.ResetSearch ();
coreRepeatSearch.ResetSearch ();
locusNameSearch.FindNextTag (0, endPos, nameString);
nextLocus = FindLocusByName (nameString);
if (nextLocus == NULL) {
cout << "Could not find locus named " << nameString.GetData () << ". Exiting..." << endl;
return -155;
}
if (!coreRepeatSearch.FindNextTag (0, endPos, repeatString))
repeatNumber = 4;
else
repeatNumber = repeatString.ConvertToInteger ();
locusInsert = "";
minSearch = nextLocus->GetMinSearchILSBP () - repeatNumber + 1;
maxSearch = nextLocus->GetMaxSearchILSBP () + repeatNumber -1;
locusInsert << "\t\t\t\t\t<Region>\n";
locusInsert << "\t\t\t\t\t\t<ILSName>" << ilsName->GetData () << "</ILSName>\n";
locusInsert << "\t\t\t\t\t\t<MinGrid>" << 0.01 * floor (100.0 * minSearch + 0.5) << "</MinGrid>\n";
locusInsert << "\t\t\t\t\t\t<MaxGrid>" << 0.01 * floor (100.0 * maxSearch + 0.5) << "</MaxGrid>\n";
locusInsert << "\t\t\t\t\t</Region>\n";
locusInsert << "\t\t\t\t</SearchRegions>";
endPos = 0;
currentLocusString.FindAndReplaceNextSubstring ("\t\t\t\t</SearchRegions>", locusInsert, endPos);
newLadderString << "\t\t\t<Locus>" << currentLocusString << "</Locus>\n";
delete newLocusString;
}
newLadderString << "\t\t</Set>\n";
newLadderString << "\t</Kits>\n";
newLadderString << "</KitData>\n";
RGString ladderPath = inFile->GetOutputConfigDirectoryPath () + "/" + inFile->GetLadderFileName ();
RGTextOutput ladderOutput (ladderPath, FALSE);
if (!ladderOutput.FileIsValid ()) {
cout << "Could not open ladder output file: " << ladderPath.GetData () << endl;
return -161;
}
ladderOutput << newLadderString;
cout << "Ladder update completed successfully..." << endl;
return 0;
}
