bool FusionCHPMultiDataAccessor::Initialize(const vector<string> &chps)
{
// Clear the map
snpNameIndexMap.clear();
// Store the chp file names.
chpFileNames = chps;
if (chps.size() == 0)
return false;
// Read the first chp file.
FusionCHPData *chp = FusionCHPDataReg::Read(chpFileNames[0]);
if (chp == NULL)
return false;
FusionCHPMultiDataData *mchp = FusionCHPMultiDataData::FromBase(chp);
if (mchp == NULL)
{
delete chp;
return false;
}
// Extract the probe set names
int n = mchp->GetEntryCount(GenotypeMultiDataType);
for (int i=0; i<n; i++)
snpNameIndexMap[mchp->GetProbeSetName(GenotypeMultiDataType, i)] = i;
// Close the file and return
delete mchp;
return true;
}
void FusionCHPMultiDataAccessor::ExtractData(const vector<string> &snps, vector<vector<u_int8_t> > &calls, vector<vector<float> > &confidences)
{
int nchps = (int)chpFileNames.size();
calls.resize(nchps);
confidences.resize(nchps);
// Create a vector of CHP file indicies given the SNP names. This is
// done to reduce the number of lookups in the map object.
int nsnps = (int)snps.size();
vector<int> snpIndicies(nsnps);
for (int isnp=0; isnp<nsnps; isnp++)
{
snpIndicies[isnp] = snpNameIndexMap[snps[isnp]];
}
// Loop over the chp files and extract the snp data
FusionCHPData *chp = NULL;
FusionCHPMultiDataData *mchp = NULL;
for (int ichp=0; ichp<nchps; ichp++)
{
// Open the CHP file
chp = FusionCHPDataReg::Read(chpFileNames[ichp]);
mchp = FusionCHPMultiDataData::FromBase(chp);
if (mchp == NULL)
{
calls[ichp].clear();
confidences[ichp].clear();
delete chp;
chp = NULL;
mchp = NULL;
continue;
}
// Extract the data
calls[ichp].resize(nsnps);
confidences[ichp].resize(nsnps);
for (int isnp=0; isnp<nsnps; isnp++)
{
calls[ichp][isnp] = mchp->GetGenoCall(GenotypeMultiDataType, snpIndicies[isnp]);
confidences[ichp][isnp] = mchp->GetGenoConfidence(GenotypeMultiDataType, snpIndicies[isnp]);
}
// Close the file
delete mchp;
}
}
void CopyNumberResultWriterTest::testWrite()
{
CopyNumberResultWriter writer;
const char *fileNames[] = {"test1.cn", "test2.cn"};
list<ParameterNameValueType> algParams;
list<ParameterNameValueType> sumParams;
vector<ColumnInfo> cols;
writer.MaximumProbeSetNameLength(12);
writer.MaximumCytoRegionNameLength(12);
writer.MaximumGenotypeProbeSetNameLength(12);
writer.AlgName() = "MYALG";
writer.AlgVersion() = "1.0";
writer.NumberProbeSets() = 10;
writer.NumberCytoRegions() = 10;
writer.NumberGenotypeProbeSets() = 10;
writer.Columns() = cols;
writer.AlgParams() = algParams;
writer.SetChromosomeProbeSetIndexInformation(X_CHR, 0, writer.NumberProbeSets());
for (int i=0; i<2; i++)
{
FusionCELData cel;
cel.SetFileName("../data/small_cel_file");
cel.Read(false);
writer.SummaryParams() = sumParams;
writer.CreateResultFile(cel, fileNames[i]);
cel.Close();
ProbeSetMultiDataCopyNumberData entry;
char buf[64];
for (int j=0; j<writer.NumberProbeSets(); j++)
{
entry.chr = X_CHR;
entry.position = j+i;
snprintf(buf, 64, "%d", j+i);
entry.name = buf;
writer.WriteProbeSetResult(entry);
}
ProbeSetMultiDataCytoRegionData cy;
for (int j=0; j<writer.NumberCytoRegions(); j++)
{
cy.call = 1;
cy.confidenceScore = (float)(j+i);
snprintf(buf, 64, "%d", j+i);
cy.name = buf;
writer.WriteCytoRegionResult(cy);
}
ProbeSetMultiDataGenotypeData gt;
for (int j=0; j<writer.NumberGenotypeProbeSets(); j++)
{
gt.call = 1;
gt.confidence = (float)(j+i);
snprintf(buf, 64, "%d", j+i);
gt.name = buf;
writer.WriteGenotypeProbeSetResult(gt);
}
writer.CloseResultsFile();
}
for (int i=0; i<2; i++)
{
FusionCHPData *chp = FusionCHPDataReg::Read(fileNames[i]);
CPPUNIT_ASSERT(chp != NULL);
FusionCHPMultiDataData *genoChp = FusionCHPMultiDataData::FromBase(chp);
CPPUNIT_ASSERT(genoChp != NULL);
CPPUNIT_ASSERT(genoChp->GetAlgName() == L"MYALG");
CPPUNIT_ASSERT(genoChp->GetAlgVersion() == L"1.0");
CPPUNIT_ASSERT(genoChp->GetArrayType() == L"Hg-small");
CPPUNIT_ASSERT(genoChp->GetEntryCount(ExpressionMultiDataType) == 0);
CPPUNIT_ASSERT(genoChp->GetEntryCount(GenotypeMultiDataType) == 10);
CPPUNIT_ASSERT(genoChp->GetEntryCount(CopyNumberMultiDataType) == 10);
CPPUNIT_ASSERT(genoChp->GetEntryCount(CytoMultiDataType) == 10);
DataSetHeader *dsh = genoChp->GetDataSetHeader(CopyNumberMultiDataType);
CPPUNIT_ASSERT(dsh->GetNameValParamCnt() == 3);
ParameterNameValueTypeConstIt begin;
ParameterNameValueTypeConstIt end;
ParameterNameValueTypeConstIt it;
dsh->GetNameValIterators(begin, end);
it = begin;
CPPUNIT_ASSERT(it->GetValueInt32() == 0);
++it;
CPPUNIT_ASSERT(it->GetValueInt32() == 10);
++it;
CPPUNIT_ASSERT(it->GetValueAscii() == "X");
ProbeSetMultiDataCopyNumberData entry;
char buf[64];
for (int j=0; j<10; j++)
{
genoChp->GetCopyNumberEntry(CopyNumberMultiDataType, j, entry);
CPPUNIT_ASSERT(entry.chr == X_CHR);
CPPUNIT_ASSERT(entry.position == j+i);
snprintf(buf, 64, "%d", j+i);
CPPUNIT_ASSERT(entry.name.compare(buf) == 0);
}
ProbeSetMultiDataCytoRegionData cy;
for (int j=0; j<10; j++)
{
genoChp->GetCytoRegionEntry(CytoMultiDataType, j, cy);
CPPUNIT_ASSERT(cy.call == 1);
CPPUNIT_ASSERT_DOUBLES_EQUAL(cy.confidenceScore, j+i, 0.0001f);
snprintf(buf, 64, "%d", j+i);
CPPUNIT_ASSERT(cy.name.compare(buf) == 0);
}
ProbeSetMultiDataGenotypeData gt;
for (int j=0; j<10; j++)
{
genoChp->GetGenotypeEntry(GenotypeMultiDataType, j, gt);
CPPUNIT_ASSERT(gt.call == 1);
CPPUNIT_ASSERT_DOUBLES_EQUAL(gt.confidence, j+i, 0.0001f);
snprintf(buf, 64, "%d", j+i);
CPPUNIT_ASSERT(gt.name.compare(buf) == 0);
}
delete chp;
}
}
