bool Locus::operator>( const Locus& rhs ) const
{
int chrComp = chr_.compare(rhs.chr());
bool result = ( chrComp > 0 )
|| ( chrComp == 0 && (this->decimalPos() > rhs.decimalPos()) );
return result;
}
void
write_plink_file_gz(const std::vector<Individual> &individuals,
const std::vector<Locus> &loci,
bool print_header,
std::string output_path,
std::auto_ptr<DoseWriter> writer)
{
gz::ogzstream output_file( output_path.c_str( ) );
output_file << std::fixed << std::setprecision( 3 );
/* Write header. */
if( print_header )
{
output_file << "SNP\tA1\tA2";
for(unsigned int i = 0; i < individuals.size( ); i++)
{
output_file << "\t" << individuals[ i ].get_fid( ) << " " << individuals[ i ].get_iid( );
}
output_file << std::endl;
}
/* Write doses */
for(unsigned int i = 0; i < loci.size( ); i++)
{
Locus locus = loci[ i ];
output_file << locus.get_name( ) << "\t" << locus.get_a1( ) << "\t" << locus.get_a2( );
for(unsigned int j = 0; j < individuals.size( ); j++)
{
writer->write( output_file, individuals[ j ], i );
}
output_file << std::endl;
}
}
GenomeTemplate BlanketResolver::resolve(
std::vector<Genome*> blanketGenomes,
std::vector<bool> & usedGenomes,
unsigned int target
) {
GenomeTemplate resolved;
GenomeTemplate unresolved = blanketGenomes[target]->getTemplate();
for (unsigned int i = 0; i < unresolved.genomeLength(); i++) {
Locus * temp = unresolved.getLocus(i);
if (!temp->isConstructive()) {
resolved.add(unresolved.getGene(i));
} else {
resolved.add(BlanketResolver::resolve(
blanketGenomes,
usedGenomes,
BlanketResolver::findMetaComponentIndex(
blanketGenomes,
usedGenomes,
((PopulationLocus*)temp)->getNode()
)
));
}
}
return resolved;
}
void Ladder :: OutputTo (RGTextOutput& xmlFile, LadderInputFile& inputFile) {
RGString link ("http://www.w3.org/2001/XMLSchema-instance");
xmlFile << "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n";
xmlFile << "<KitData xmlns:xsi=\"" << link.GetData () << "\" xsi:noNamespaceSchemaLocation=\"MarkerSet.xsd\">\n";
xmlFile << "\t<Version>2.0</Version>\n";
xmlFile << "\t<Kits>\n";
xmlFile << "\t\t<Set>\n";
xmlFile << "\t\t\t<Name>" << mMarkerSetName.GetData () << "</Name>\n";
xmlFile << "\t\t\t<NChannels>" << mNumberOfChannels << "</NChannels>\n";
xmlFile << "\t\t\t<LS>\n";
OutputILSListTo (xmlFile);
xmlFile << "\t\t\t\t<ChannelNo>" << mChannelForILS << "</ChannelNo>\n";
xmlFile << "\t\t\t</LS>\n";
xmlFile << "\t\t\t<FileNameSuffix>" << mSuffix << "</FileNameSuffix>\n";
xmlFile << "\t\t\t<GenotypeSuffix>" << mSuffix << "</GenotypeSuffix>\n";
xmlFile << "\t\t\t<DirectorySearchString>" << mSuffix << "</DirectorySearchString>\n";
OutputChannelMapTo (xmlFile, inputFile);
list <Locus*>::const_iterator locusIterator;
Locus* nextLocus;
for (locusIterator = mLocusList.begin(); locusIterator != mLocusList.end(); locusIterator++) {
nextLocus = *locusIterator;
nextLocus->OutputTo (xmlFile);
}
xmlFile << "\t\t</Set>\n";
xmlFile << "\t</Kits>\n";
xmlFile << "</KitData>";
}
int Ladder :: MergeThisLadderInto (Ladder* ladder) {
list <Locus*>::const_iterator locusIterator;
Locus* nextLocus;
Locus* prevLocus = NULL;
double prevMax;
double nextMin;
double nextOriginal;
double prevOriginal;
for (locusIterator = mLocusList.begin(); locusIterator != mLocusList.end(); locusIterator++) {
nextLocus = *locusIterator;
ladder->MergeLocusIntoLadder (nextLocus);
}
// To do: test that search regions do not overlap neighboring loci
for (locusIterator = mLocusList.begin(); locusIterator != mLocusList.end(); locusIterator++) {
nextLocus = *locusIterator;
if (prevLocus == NULL) {
prevLocus = nextLocus;
continue;
}
if (nextLocus->GetChannel () != prevLocus->GetChannel ()) {
prevLocus = nextLocus;
continue;
}
prevMax = prevLocus->GetMaxSearchILSBP ();
nextMin = nextLocus->GetMinSearchILSBP ();
if (prevMax <= nextMin) {
prevLocus = nextLocus;
continue;
}
nextOriginal = nextLocus->GetOriginalMinSearchILSBP () - 0.55;
prevOriginal = prevLocus->GetOriginalMaxSearchILSBP () + 0.55;
if (prevOriginal > nextOriginal)
prevOriginal = nextOriginal = 0.5 * (prevOriginal + nextOriginal);
if (prevMax >= nextOriginal)
prevLocus->SetMaxSearhILSBP (nextOriginal);
if (nextMin <= prevOriginal)
nextLocus->SetMinSearhILSBP (prevOriginal);
prevLocus = nextLocus;
}
return 0;
}
static void
update_dmr_name(Locus &dmr, double lod_sum, double mindiff_sum) {
ostringstream name_stream;
name_stream << "dmr:"
<< lod_sum / dmr.score() << ":"
<< mindiff_sum / dmr.score();
dmr.set_name(name_stream.str());
}
bool Ladder :: TestAllLociMerged () {
list <Locus*>::const_iterator locusIterator;
Locus* nextLocus;
for (locusIterator = mLocusList.begin(); locusIterator != mLocusList.end(); locusIterator++) {
nextLocus = *locusIterator;
if (!nextLocus->GetMerged ())
return false;
}
return true;
}
Locus* Ladder :: FindLocusByName (const RGString& locusName) {
Locus* nextLocus;
list <Locus*>::const_iterator locusIterator;
for (locusIterator = mLocusList.begin(); locusIterator != mLocusList.end(); locusIterator++) {
nextLocus = *locusIterator;
if (nextLocus->GetName () == locusName)
return nextLocus;
}
return NULL;
}
Gene* GaussianMutation::newLocusValue(Gene* current) {
double addend = HierRNG::gaussian(this->mean, this->stdDev);
double newIndex = current->getIndex() + addend;
Locus* locus = current->getLocus();
if (!this->endReflection)
return current->copy(locus->closestIndex(newIndex));
while (locus->outOfRange(newIndex)) {
double topIndex = locus->topIndex();
double bottomIndex = locus->bottomIndex();
if (newIndex > topIndex) {
newIndex = topIndex - (newIndex - topIndex);
} else if (newIndex < bottomIndex) {
newIndex = bottomIndex - (newIndex - bottomIndex);
}
}
return current->copy(newIndex);
}
int Ladder :: AddLocus (Locus* newLocus) {
list <Locus*>::const_iterator locusIterator;
Locus* nextLocus;
int status = 0;
for (locusIterator = mLocusList.begin(); locusIterator != mLocusList.end(); locusIterator++) {
nextLocus = *locusIterator;
if (nextLocus->isEqual (newLocus)) {
status = -1;
break;
}
}
if (status == 0)
mLocusList.push_back (newLocus);
return status;
}
Ladder* Bins :: AssembleAllLoci (RGDList& doNotExtend) {
int nLoci;
Ladder* newLadder = new Ladder ();
Locus* nextLocus;
nLoci = SplitFileIntoLocusStrings ();
int i;
int status;
RGString name;
if (nLoci <= 0) {
cout << "Could not complete ladder setup" << endl;
delete newLadder;
return NULL;
}
for (i=1; i<=nLoci; i++) {
SplitLocusStringIntoLines (i);
nextLocus = AssembleLocusFromLineStrings ();
status = newLadder->AddLocus (nextLocus);
if (status < 0) {
cout << "Locus name repeated in bins file: " << nextLocus->GetName ().GetData () << endl;
delete newLadder;
return NULL;
}
name = nextLocus->GetName ();
if (doNotExtend.Contains (&name))
nextLocus->SetDoNotExtend ();
}
return newLadder;
}
int Ladder :: MergeLocusIntoLadder (const Locus* locus) {
// This combines info and then computes the rest
Locus* matchingLocus = FindLocusByName (locus->GetName ());
if (matchingLocus == NULL) {
cout << "Could not find locus matching name: " << locus->GetName ().GetData () << endl;
return -1;
}
matchingLocus->SetFirstCoreLocusBP (locus->GetFirstCoreLocusBP ());
matchingLocus->SetFirstExtendedAllele (locus->GetFirstExtendedAllele ());
matchingLocus->SetLastExtendedAllele (locus->GetLastExtendedAllele ());
matchingLocus->ComputeAllBPs ();
matchingLocus->SetMinMaxSearchILSBP (locus->GetMinSearchILSBP (), locus->GetMaxSearchILSBP ());
matchingLocus->AdjustSearchRegion ();
matchingLocus->SetMerged ();
return 0;
}
int CoreBioComponent :: InitializeSM (SampleData& fileData, PopulationCollection* collection, const RGString& markerSetName, Boolean isGrid) {
//
// This is ladder and sample stage 1
//
mTime = fileData.GetCollectionStartTime ();
mDate = fileData.GetCollectionStartDate ();
mName = fileData.GetName ();
mRunStart = mDate.GetOARString () + mTime.GetOARString ();
mMarkerSet = collection->GetNamedPopulationMarkerSet (markerSetName);
Progress = 0;
smMarkerSetNameUnknown noNamedMarkerSet;
smNamedILSUnknown noNamedILS;
smChannelIsILS channelIsILS;
if (mMarkerSet == NULL) {
ErrorString = "*******COULD NOT FIND MARKER SET NAMED ";
ErrorString << markerSetName << " IN POPULATION COLLECTION********\n";
SetMessageValue (noNamedMarkerSet, true);
AppendDataForSmartMessage (noNamedMarkerSet, markerSetName);
return -1;
}
mLaneStandard = mMarkerSet->GetLaneStandard ();
mNumberOfChannels = mMarkerSet->GetNumberOfChannels ();
if ((mLaneStandard == NULL) || !mLaneStandard->IsValid ()) {
ErrorString = "Could not find named internal lane standard associated with marker set named ";
ErrorString << markerSetName << "\n";
cout << "Could not find named internal lane standard associated with marker set named " << (char*)markerSetName.GetData () << endl;
SetMessageValue (noNamedILS, true);
AppendDataForSmartMessage (noNamedILS, markerSetName);
return -1;
}
mDataChannels = new ChannelData* [mNumberOfChannels + 1];
int i;
const int* fsaChannelMap = mMarkerSet->GetChannelMap ();
for (i=0; i<=mNumberOfChannels; i++)
mDataChannels [i] = NULL;
mLaneStandardChannel = mMarkerSet->GetLaneStandardChannel ();
for (i=1; i<=mNumberOfChannels; i++) {
if (i == mLaneStandardChannel)
mDataChannels [i] = GetNewLaneStandardChannel (i, mLaneStandard);
else {
if (isGrid)
mDataChannels [i] = GetNewGridDataChannel (i, mLaneStandard);
else
mDataChannels [i] = GetNewDataChannel (i, mLaneStandard);
}
mDataChannels [i]->SetFsaChannel (fsaChannelMap [i]);
}
mLSData = mDataChannels [mLaneStandardChannel];
mLSData->SetMessageValue (channelIsILS, true);
mMarkerSet->ResetLocusList ();
Locus* nextLocus;
while (nextLocus = mMarkerSet->GetNextLocus ())
mDataChannels [nextLocus->GetLocusChannel ()]->AddLocus (nextLocus);
Progress = 1;
return 0;
}
void CoreBioComponent :: ReportXMLSmartSampleTableRowWithLinks (RGTextOutput& text, RGTextOutput& tempText) {
RGString type;
if (mIsNegativeControl)
type = "-Control";
else if (mIsPositiveControl)
type = "+Control";
else
type = "Sample";
RGString pResult;
RGString SimpleFileName (mName);
size_t startPos = 0;
size_t endPos;
size_t length = SimpleFileName.Length ();
if (SimpleFileName.FindLastSubstringCaseIndependent (DirectoryManager::GetDataFileType (), startPos, endPos)) {
if (endPos == length - 1)
SimpleFileName.ExtractAndRemoveLastCharacters (4);
}
SimpleFileName.FindAndReplaceAllSubstrings ("\\", "/");
startPos = endPos = 0;
if (SimpleFileName.FindLastSubstring ("/", startPos, endPos)) {
SimpleFileName.ExtractAndRemoveSubstring (0, startPos);
}
text << CLevel (1) << "\t\t<Sample>\n";
text << "\t\t\t<Name>" << xmlwriter::EscAscii (SimpleFileName, &pResult) << "</Name>\n";
text << "\t\t\t<SampleName>" << xmlwriter::EscAscii (mSampleName, &pResult) << "</SampleName>\n";
text << "\t\t\t<RunStart>" << mRunStart.GetData () << "</RunStart>\n";
text << "\t\t\t<Type>" << type.GetData () << "</Type>\n" << PLevel ();
int trigger = Notice::GetMessageTrigger ();
// int channelHighestLevel;
// bool channelAlerts = false;
int cbcHighestMsgLevel = GetHighestMessageLevelWithRestrictionSM ();
RGDListIterator it (*mSmartMessageReporters);
SmartMessageReporter* nextNotice;
bool includesExportInfo = false;
while (nextNotice = (SmartMessageReporter*) it ()) {
if (nextNotice->HasViableExportInfo ()) {
includesExportInfo = true;
break;
}
}
if (((cbcHighestMsgLevel > 0) && (cbcHighestMsgLevel <= trigger)) || includesExportInfo) {
// text << CLevel (1) << "\t\t\t<SampleAlerts>\n" << PLevel ();
// get message numbers and report
ReportXMLSmartNoticeObjects (text, tempText, " ");
// text << CLevel (1) << "\t\t\t</SampleAlerts>\n" << PLevel ();
}
mDataChannels [mLaneStandardChannel]->ReportXMLILSSmartNoticeObjects (text, tempText, " ");
int i;
//for (i=1; i<=mNumberOfChannels; i++) {
// //if (i == mLaneStandardChannel)
// // continue;
// channelHighestLevel = mDataChannels [i]->GetHighestMessageLevelWithRestrictionSM ();
// if ((channelHighestLevel > 0) && (channelHighestLevel <= trigger)) {
// channelAlerts = true;
// break;
// }
//}
// if (channelAlerts) {
text << CLevel (1) << "\t\t\t<ChannelAlerts>\n" << PLevel ();
for (i=1; i<=mNumberOfChannels; i++) {
if (i == mLaneStandardChannel)
continue;
mDataChannels [i]->ReportXMLSmartNoticeObjects (text, tempText, " ");
}
text << CLevel (1) << "\t\t\t</ChannelAlerts>\n" << PLevel ();
// }
mMarkerSet->ResetLocusList ();
Locus* nextLocus;
while (nextLocus = mMarkerSet->GetNextLocus ()) {
nextLocus->ReportXMLSmartSampleTableRowWithLinks (text, tempText, " ");
}
if (mIsPositiveControl)
text << CLevel (1) << "\t\t\t<PositiveControl>" << mPositiveControlName << "</PositiveControl>\n";
text << CLevel (1) << "\t\t</Sample>\n" << PLevel ();
}
Locus* Bins :: AssembleLocusFromLineStrings () {
// uses current values in mLocusLineStrings
Locus* newLocus = new Locus ();
int nLines = mLocusLineStrings.Length ();
SplitLocusLine (mLocusLineStrings [0]); // This should give us the name field
RGString name = mLocusLineFieldString [1];
cout << "Locus name = " << name.GetData () << endl;
//cout << "Bins locus line 1: ";
int i;
int len = mLocusLineFieldString.Length ();
//for (i=0; i<len; i++) {
// cout << mLocusLineFieldString [i].GetData ();
// if (i < len - 1)
// cout << ", ";
//}
//cout << endl;
RGString firstAlleleName;
int firstLocusBP;
double firstILSBP;
double lastILSBP;
RGString firstExtendedAlleleName;
RGString lastExtendedAlleleName;
int lineForFirstAllele = 0;
for (i=1; i<nLines; i++) {
SplitLocusLine (mLocusLineStrings [i]);
if (mLocusLineFieldString.Length () < 4) {
cout << "Could not find first allele for locus " << name.GetData () << endl;
delete newLocus;
return NULL;
}
if (mLocusLineFieldString.Length () >= 5) {
if (mLocusLineFieldString [4] == "virtual") {
if (i == 1) {
firstExtendedAlleleName = mLocusLineFieldString [0];
}
continue;
}
}
firstAlleleName = mLocusLineFieldString [0];
firstILSBP = mLocusLineFieldString [1].ConvertToDouble ();
firstLocusBP = (int) floor (firstILSBP + 0.5);
lineForFirstAllele = i;
break;
}
if (firstExtendedAlleleName.Length () == 0)
firstExtendedAlleleName = firstAlleleName;
bool foundLastExtendedAllele = false;
for (i=nLines-1; i>lineForFirstAllele; i--) {
SplitLocusLine (mLocusLineStrings [i]);
if (mLocusLineFieldString.Length () < 4)
continue;
if (!foundLastExtendedAllele) {
lastExtendedAlleleName = mLocusLineFieldString [0];
foundLastExtendedAllele = true;
}
if ((mLocusLineFieldString.Length () >= 5) && (mLocusLineFieldString [4] == "virtual"))
continue;
lastILSBP = mLocusLineFieldString [1].ConvertToDouble ();
break;
}
if (lastExtendedAlleleName.Length () == 0) {
cout << "Could not locate last extended allele name for locus: " << name.GetData () << endl;
delete newLocus;
return NULL;
}
newLocus->SetFirstCoreAllele (firstAlleleName);
newLocus->SetFirstExtendedAllele (firstExtendedAlleleName);
newLocus->SetLastExtendedAllele (lastExtendedAlleleName);
newLocus->SetMinMaxSearchILSBP (firstILSBP, lastILSBP); // These will have to be adjusted later when core repeat number is known
newLocus->SetFirstCoreLocusBP (firstLocusBP);
newLocus->SetName (name);
// Need to add to Locus ability to compute first and last extended bp
return newLocus;
}
void
merge(istream &cpg_stream, ostream &dmr_stream, double cutoff) {
bool skipped_last_cpg;
Locus dmr;
if (!read_next_significant_cpg(cpg_stream, dmr, cutoff, skipped_last_cpg))
return;
dmr.set_score(1);
double lod_sum, mindiff_sum, unadjusted_pval;
extract_scores(dmr.name(), lod_sum, mindiff_sum, unadjusted_pval);
if (unadjusted_pval >= cutoff) {
lod_sum = 0;
mindiff_sum = 0;
}
Locus cpg;
while(read_next_significant_cpg(cpg_stream, cpg, cutoff, skipped_last_cpg)) {
double lod, mindiff;
extract_scores(cpg.name(), lod, mindiff, unadjusted_pval);
if (skipped_last_cpg || cpg.chrom() != dmr.chrom()) {
update_dmr_name(dmr, lod_sum, mindiff_sum);
if (lod_sum != 0) {
dmr_stream << dmr << std::endl;
}
dmr = cpg;
dmr.set_score(1);
if (unadjusted_pval >= cutoff) {
lod_sum = 0;
mindiff_sum = 0;
} else {
lod_sum = lod;
mindiff_sum = mindiff;
}
} else {
dmr.set_end(cpg.end());
dmr.set_score(dmr.score() + 1);
if (unadjusted_pval < cutoff) {
lod_sum += lod;
mindiff_sum += mindiff;
}
}
}
update_dmr_name(dmr, lod_sum, mindiff_sum);
dmr_stream << dmr << std::endl;
}
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;
}