void CHPMultiDataDataTest::test_SumParams()
{
ParameterNameValueTypeList params;
ParameterNameValueType param;
CHPMultiDataData data;
param.SetName(L"n1");
param.SetValueText(L"v1");
params.push_back(param);
param.SetName(L"n2");
param.SetValueText(L"v2");
params.push_back(param);
data.AddSummaryParams(params);
ParameterNameValueTypeList params_out = data.GetSummaryParams();
CPPUNIT_ASSERT(params_out.size() == 2);
ParameterNameValueTypeList::iterator it = params_out.begin();
param = *it;
CPPUNIT_ASSERT(param.GetName() == L"n1");
CPPUNIT_ASSERT(param.GetValueText() == L"v1");
++it;
param = *it;
CPPUNIT_ASSERT(param.GetName() == L"n2");
CPPUNIT_ASSERT(param.GetValueText() == L"v2");
}
void CalvinCHPMultiDataFileUpdaterTest::testMultiData1()
{
CreateReferenceFile1();
vector<ColumnInfo> cols;
CalvinCHPMultiDataFileUpdater upd;
ProbeSetMultiDataGenotypeData e;
upd.Initialize(TEST1_FILE);
e.call = 11;
e.confidence = 111.0f;
upd.UpdateMultiData(GenotypeMultiDataType, 0, e);
e.call = 22;
e.confidence = 222.0f;
upd.UpdateMultiData(GenotypeMultiDataType, 2, e);
CHPMultiDataData data;
CHPMultiDataFileReader reader;
CPPUNIT_ASSERT_NO_THROW(reader.SetFilename(TEST1_FILE));
CPPUNIT_ASSERT_NO_THROW(reader.Read(data));
CPPUNIT_ASSERT(data.GetEntryCount(GenotypeMultiDataType) == 4);
data.GetGenotypeEntry(GenotypeMultiDataType, 0, e);
CPPUNIT_ASSERT(e.name == "1");
CPPUNIT_ASSERT(e.call == 11);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.confidence, 111.0f, 0.000001f);
data.GetGenotypeEntry(GenotypeMultiDataType, 1, e);
CPPUNIT_ASSERT(e.name == "2");
CPPUNIT_ASSERT(e.call == 2);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.confidence, 20.0f, 0.000001f);
data.GetGenotypeEntry(GenotypeMultiDataType, 2, e);
CPPUNIT_ASSERT(e.name == "3");
CPPUNIT_ASSERT(e.call == 22);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.confidence, 222.0f, 0.000001f);
data.GetGenotypeEntry(GenotypeMultiDataType, 3, e);
CPPUNIT_ASSERT(e.name == "4");
CPPUNIT_ASSERT(e.call == 4);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.confidence, 40.0f, 0.000001f);
}
void CalvinCHPMultiDataFileUpdaterTest::testMultiData3()
{
CreateReferenceFile3();
vector<ColumnInfo> cols;
ByteColumn bcol(L"byte");
cols.push_back(bcol);
UByteColumn ubcol(L"ubyte");
cols.push_back(ubcol);
ShortColumn scol(L"short");
cols.push_back(scol);
UShortColumn uscol(L"ushort");
cols.push_back(uscol);
IntColumn icol(L"int");
cols.push_back(icol);
UIntColumn uicol(L"uint");
cols.push_back(uicol);
FloatColumn fcol(L"float");
cols.push_back(fcol);
ASCIIColumn acol(L"ascii", 7);
cols.push_back(acol);
UnicodeColumn tcol(L"text", 10);
cols.push_back(tcol);
CalvinCHPMultiDataFileUpdater upd;
ProbeSetMultiDataCopyNumberData e;
ProbeSetMultiDataCytoRegionData c;
upd.Initialize(TEST3_FILE);
e.metrics.resize(9);
e.chr = 11;
e.position = 111;
e.metrics[0].SetValueInt8(9);
e.metrics[1].SetValueUInt8(10);
e.metrics[2].SetValueInt16(17);
e.metrics[3].SetValueUInt16(18);
e.metrics[4].SetValueInt32(33);
e.metrics[5].SetValueUInt32(34);
e.metrics[6].SetValueFloat(55.0f);
e.metrics[7].SetValueAscii("text");
e.metrics[8].SetValueText(L"ascii");
upd.UpdateMultiData(CopyNumberMultiDataType, 0, e, cols);
e.chr = 22;
e.position = 222;
e.metrics[0].SetValueInt8(10);
e.metrics[1].SetValueUInt8(11);
e.metrics[2].SetValueInt16(18);
e.metrics[3].SetValueUInt16(19);
e.metrics[4].SetValueInt32(34);
e.metrics[5].SetValueUInt32(35);
e.metrics[6].SetValueFloat(66.0f);
e.metrics[7].SetValueAscii("text2");
e.metrics[8].SetValueText(L"ascii2");
upd.UpdateMultiData(CopyNumberMultiDataType, 2, e, cols);
vector<ColumnInfo> cycols;
c.call = 11;
c.confidenceScore = 111.0f;
c.chr = 11;
c.startPosition = 11;
c.stopPosition = 22;
upd.UpdateMultiData(CytoMultiDataType, 0, c, cycols);
c.call = 2;
c.confidenceScore = 20.0f;
c.chr = 22;
c.startPosition = 22;
c.stopPosition = 33;
upd.UpdateMultiData(CytoMultiDataType, 1, c, cycols);
CHPMultiDataData data;
CHPMultiDataFileReader reader;
CPPUNIT_ASSERT_NO_THROW(reader.SetFilename(TEST3_FILE));
CPPUNIT_ASSERT_NO_THROW(reader.Read(data));
CPPUNIT_ASSERT(data.GetEntryCount(CopyNumberMultiDataType) == 4);
CPPUNIT_ASSERT(data.GetEntryCount(CytoMultiDataType) == 2);
data.GetCopyNumberEntry(CopyNumberMultiDataType, 0, e);
CPPUNIT_ASSERT(e.name == "1");
CPPUNIT_ASSERT(e.chr == 11);
CPPUNIT_ASSERT(e.position == 111);
CPPUNIT_ASSERT(e.metrics[0].GetValueInt8() == 9);
CPPUNIT_ASSERT(e.metrics[1].GetValueUInt8() == 10);
CPPUNIT_ASSERT(e.metrics[2].GetValueInt16() == 17);
CPPUNIT_ASSERT(e.metrics[3].GetValueUInt16() == 18);
CPPUNIT_ASSERT(e.metrics[4].GetValueInt32() == 33);
CPPUNIT_ASSERT(e.metrics[5].GetValueUInt32() == 34);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.metrics[6].GetValueFloat(), 55.0f, 0.00001f);
CPPUNIT_ASSERT(e.metrics[7].GetValueAscii() == "text");
CPPUNIT_ASSERT(e.metrics[8].GetValueText() == L"ascii");
data.GetCopyNumberEntry(CopyNumberMultiDataType, 1, e);
CPPUNIT_ASSERT(e.name == "2");
CPPUNIT_ASSERT(e.chr == 2);
CPPUNIT_ASSERT(e.position == 20);
CPPUNIT_ASSERT(e.metrics[0].GetValueInt8() == 8);
CPPUNIT_ASSERT(e.metrics[1].GetValueUInt8() == 8);
CPPUNIT_ASSERT(e.metrics[2].GetValueInt16() == 16);
CPPUNIT_ASSERT(e.metrics[3].GetValueUInt16() == 16);
CPPUNIT_ASSERT(e.metrics[4].GetValueInt32() == 32);
CPPUNIT_ASSERT(e.metrics[5].GetValueUInt32() == 32);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.metrics[6].GetValueFloat(), 44.0f, 0.00001f);
CPPUNIT_ASSERT(e.metrics[7].GetValueAscii() == "ascii");
CPPUNIT_ASSERT(e.metrics[8].GetValueText() == L"text");
data.GetCopyNumberEntry(CopyNumberMultiDataType, 2, e);
CPPUNIT_ASSERT(e.name == "3");
CPPUNIT_ASSERT(e.chr == 22);
CPPUNIT_ASSERT(e.position == 222);
CPPUNIT_ASSERT(e.metrics[0].GetValueInt8() == 10);
CPPUNIT_ASSERT(e.metrics[1].GetValueUInt8() == 11);
CPPUNIT_ASSERT(e.metrics[2].GetValueInt16() == 18);
CPPUNIT_ASSERT(e.metrics[3].GetValueUInt16() == 19);
CPPUNIT_ASSERT(e.metrics[4].GetValueInt32() == 34);
CPPUNIT_ASSERT(e.metrics[5].GetValueUInt32() == 35);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.metrics[6].GetValueFloat(), 66.0f, 0.00001f);
CPPUNIT_ASSERT(e.metrics[7].GetValueAscii() == "text2");
CPPUNIT_ASSERT(e.metrics[8].GetValueText() == L"ascii2");
data.GetCopyNumberEntry(CopyNumberMultiDataType, 3, e);
CPPUNIT_ASSERT(e.name == "4");
CPPUNIT_ASSERT(e.chr == 4);
CPPUNIT_ASSERT(e.position == 40);
CPPUNIT_ASSERT(e.metrics[0].GetValueInt8() == 8);
CPPUNIT_ASSERT(e.metrics[1].GetValueUInt8() == 8);
CPPUNIT_ASSERT(e.metrics[2].GetValueInt16() == 16);
CPPUNIT_ASSERT(e.metrics[3].GetValueUInt16() == 16);
CPPUNIT_ASSERT(e.metrics[4].GetValueInt32() == 32);
CPPUNIT_ASSERT(e.metrics[5].GetValueUInt32() == 32);
CPPUNIT_ASSERT_DOUBLES_EQUAL(e.metrics[6].GetValueFloat(), 44.0f, 0.00001f);
CPPUNIT_ASSERT(e.metrics[7].GetValueAscii() == "ascii");
CPPUNIT_ASSERT(e.metrics[8].GetValueText() == L"text");
data.GetCytoEntry(CytoMultiDataType, 0, c);
CPPUNIT_ASSERT(c.name == "1");
CPPUNIT_ASSERT(c.call == 11);
CPPUNIT_ASSERT(c.chr == 11);
CPPUNIT_ASSERT(c.startPosition == 11);
CPPUNIT_ASSERT(c.stopPosition == 22);
CPPUNIT_ASSERT_DOUBLES_EQUAL(c.confidenceScore, 111.0f, 0.00001f);
data.GetCytoEntry(CytoMultiDataType, 1, c);
CPPUNIT_ASSERT(c.name == "2");
CPPUNIT_ASSERT(c.chr == 22);
CPPUNIT_ASSERT(c.startPosition == 22);
CPPUNIT_ASSERT(c.stopPosition == 33);
CPPUNIT_ASSERT(c.call == 2);
CPPUNIT_ASSERT_DOUBLES_EQUAL(c.confidenceScore, 20.0f, 0.00001f);
}
/*! Create a "multi-data" CHP file with just the header information. The remainder of the file
* will be created at a later time using the buffer writer technique.
* The CHP file will contain only "genotyping" results.
* @param execId The execution identifier. This identifier is used to identify the batch run that created the CHP files.
* @param celFile The full path to the parent CEL file. The header of the CEL file is copied to the CHP file.
* @param outFile The name of the output CHP file.
* @param extraColNames The names of the extra data columns. Should not include probe set name, call and confidence columns.
* @param extraColTypes The types (float, int, ubyte) of the extra columns.
* @param numEntries The number of rows (entries) of results to store in the CHP file.
* @param maxProbeSetNameLength The maximum length of the probe set names.
* @param algName The name of the algorithm used to create the results.
* @param algVersion The algorithm version.
* @param chipType the chip type, also known as the probe array type.
* @param programName The name of the program used to create the CHP file.
* @param programVersion The version of the program.
* @param programCompany The company or institution who developed the CHP creating software.
* @param paramNames A list of parameter names to store in the CHP file header.
* @param paramValues A list of parameter values to store in the CHP file header.
* @param sumNames A list of summary statistic names to store in the CHP file header.
* @param sumValues A list of summary statistic values to store in the CHP file header.
*/
static void CreateFileWithHeader
(
const string &execId,
const string &celFile,
const string &outFile,
const vector<string>& extraColNames,
const vector<string>& extraColTypes,
unsigned long numEntries,
int maxProbeSetNameLength,
const string &algName,
const string &algVersion,
const string &chipType,
const string &programName,
const string &programVersion,
const string &programCompany,
const vector<string>& paramNames,
const vector<string>& paramValues,
const vector<string>& sumNames,
const vector<string>& sumValues,
const vector<string>& extraNames,
const vector<string>& extraValues
)
{
// Create the vector of extra columns. The sample code here supports only float, 32 bit integers and 8 bit unsigned integers.
vector<ColumnInfo> extraColumns;
int ncols = (int)extraColNames.size();
for (int icol=0; icol<ncols; icol++)
{
if (extraColTypes[icol] == "float")
{
FloatColumn fcol(StringUtils::ConvertMBSToWCS(extraColNames[icol]));
extraColumns.push_back(fcol);
}
else if (extraColTypes[icol] == "int")
{
IntColumn intcol(StringUtils::ConvertMBSToWCS(extraColNames[icol]));
extraColumns.push_back(intcol);
}
else if (extraColTypes[icol] == "ubyte")
{
UByteColumn ubcol(StringUtils::ConvertMBSToWCS(extraColNames[icol]));
extraColumns.push_back(ubcol);
}
else
{
throw string("Unsupported column type: ") + extraColTypes[icol];
}
}
// Create the data object
CHPMultiDataData *data = new CHPMultiDataData(outFile);
data->SetEntryCount(GenotypeMultiDataType, numEntries, maxProbeSetNameLength, extraColumns);
data->SetAlgName(StringUtils::ConvertMBSToWCS(algName));
data->SetAlgVersion(StringUtils::ConvertMBSToWCS(algVersion));
data->SetArrayType(StringUtils::ConvertMBSToWCS(chipType));
// Store the CEL header
if (celFile.length() > 0 && FileUtils::Exists(celFile.c_str()) == true)
{
FusionCELData cel;
cel.SetFileName(celFile.c_str());
cel.ReadHeader();
GenericData *gdata = cel.GetGenericData();
if (gdata != NULL)
data->GetFileHeader()->GetGenericDataHdr()->AddParent(*gdata->Header().GetGenericDataHdr());
cel.Close();
}
// Add algorithm parameters to list.
ParameterNameValueTypeList params;
ParameterNameValueType param;
if (programName.empty() == false)
{
param.SetName(L"program-name");
param.SetValueText(StringUtils::ConvertMBSToWCS(programName));
data->GetGenericData().Header().GetGenericDataHdr()->AddNameValParam(param);
}
if (programVersion.empty() == false)
{
param.SetName(L"program-version");
param.SetValueText(StringUtils::ConvertMBSToWCS(programVersion));
data->GetGenericData().Header().GetGenericDataHdr()->AddNameValParam(param);
}
if (programCompany.empty() == false)
{
param.SetName(L"program-company");
param.SetValueText(StringUtils::ConvertMBSToWCS(programCompany));
data->GetGenericData().Header().GetGenericDataHdr()->AddNameValParam(param);
}
int nparams = (int) extraNames.size();
for(int iparam=0; iparam<nparams; iparam++)
{
param.SetName(StringUtils::ConvertMBSToWCS(extraNames[iparam]));
param.SetValueAscii(extraValues[iparam]);
data->GetGenericData().Header().GetGenericDataHdr()->AddNameValParam(param);
}
nparams = (int) paramNames.size();
param.SetName(L"exec-guid");
param.SetValueAscii(execId);
params.push_back(param);
for(int iparam=0; iparam<nparams; iparam++)
{
param.SetName(StringUtils::ConvertMBSToWCS(paramNames[iparam]));
param.SetValueAscii(paramValues[iparam]);
params.push_back(param);
}
if (params.empty() == false)
data->AddAlgParams(params);
params.clear();
nparams = (int) sumNames.size();
for(int iparam=0; iparam<nparams; iparam++)
{
param.SetName(StringUtils::ConvertMBSToWCS(sumNames[iparam]));
param.SetValueAscii(sumValues[iparam]);
params.push_back(param);
}
if (params.empty() == false)
data->AddSummaryParams(params);
// Creating the writer object will create the file with the header information.
CHPMultiDataFileWriter writer(*data);
}
void CHPMultiDataDataTest::test_AlgVersion()
{
CHPMultiDataData data;
data.SetAlgVersion(L"1.0");
CPPUNIT_ASSERT(data.GetAlgVersion() == L"1.0");
}
void CHPMultiDataDataTest::test_AlgName()
{
CHPMultiDataData data;
data.SetAlgName(L"alg");
CPPUNIT_ASSERT(data.GetAlgName() == L"alg");
}
void CHPMultiDataDataTest::test_ArrayType()
{
CHPMultiDataData data;
data.SetArrayType(L"test3");
CPPUNIT_ASSERT(data.GetArrayType() == L"test3");
}
void CHPMultiDataDataTest::test_FileName()
{
CHPMultiDataData data;
data.SetFilename("file");
CPPUNIT_ASSERT( data.GetFilename() == "file");
}
/*
* Create a results file with the CEL file header and other parameters.
*/
void CopyNumberResultWriter::CreateResultFile(affymetrix_fusion_io::FusionCELData& cel, const std::string& fileName)
{
try
{
// Create the results file with the header.
CHPMultiDataData *data = new CHPMultiDataData(fileName);
data->SetEntryCount(CopyNumberMultiDataType, numberProbeSets, maxProbeSetNameLength[CopyNumberMultiDataType], columns);
if (numberCytoRegions > 0)
data->SetEntryCount(CytoMultiDataType, numberCytoRegions, maxProbeSetNameLength[CytoMultiDataType],cytoRegionColumns);
if (numberGenotypeProbeSets > 0)
data->SetEntryCount(GenotypeMultiDataType, numberGenotypeProbeSets, maxProbeSetNameLength[GenotypeMultiDataType], genotypeColumns);
data->SetAlgName(StringUtils::ConvertMBSToWCS(algName));
data->SetAlgVersion(StringUtils::ConvertMBSToWCS(algVersion));
data->SetArrayType(cel.GetChipType());
GenericDataHeader *gdh = data->GetFileHeader()->GetGenericDataHdr();
ParameterNameValueType param;
param.SetName(PROGRAM_NAME);
param.SetValueText(StringUtils::ConvertMBSToWCS(programName));
gdh->AddNameValParam(param);
param.SetName(L"program-version");
param.SetValueText(StringUtils::ConvertMBSToWCS(programVersion));
gdh->AddNameValParam(param);
param.SetName(PROGRAM_COMPANY);
param.SetValueText(StringUtils::ConvertMBSToWCS(programCompany));
gdh->AddNameValParam(param);
ParameterNameValueTypeList params = algParams;
param.SetName(L"ArraySet");
param.SetValueText(cel.GetChipType());
params.push_back(param);
data->AddAlgParams(params);
data->AddSummaryParams(summaryParams);
DataSetHeader *dsh = data->GetDataSetHeader(CopyNumberMultiDataType);
for (ParameterNameValueTypeList::iterator it=chrStartStop.begin(); it!=chrStartStop.end(); it++)
dsh->AddNameValParam(*it);
GenericData *gdata = cel.GetGenericData();
if (gdata != NULL)
gdh->AddParent(*gdata->Header().GetGenericDataHdr());
CHPMultiDataFileWriter *writer = new CHPMultiDataFileWriter(*data);
delete writer;
delete data;
// Create a buffer writer object
outputFiles.clear();
outputFiles.push_back(fileName);
vector<MultiDataType> dataTypes;
dataTypes.push_back(CopyNumberMultiDataType);
if (numberCytoRegions > 0)
dataTypes.push_back(CytoMultiDataType);
if (numberGenotypeProbeSets > 0)
dataTypes.push_back(GenotypeMultiDataType);
bufferWriter = new CHPMultiDataFileBufferWriter();
bufferWriter->Initialize(&outputFiles, dataTypes, maxProbeSetNameLength);
}
catch (CalvinException &ex)
{
string err = "Error creating the output file: " + fileName;
wstring msg = ex.ToString();
if (msg.empty() == false)
err += " " + StringUtils::ConvertWCSToMBS(msg);
throw err;
}
catch (...)
{
string err = "Error creating the output file: " + fileName;
throw err;
}
}