bool VcfRecord::read(IFILE filePtr, bool siteOnly,
VcfRecordDiscardRules& discardRules,
VcfSubsetSamples* sampleSubset)
{
// Clear out any previously set values.
reset();
if(filePtr == NULL)
{
myStatus.setStatus(StatGenStatus::FAIL_ORDER,
"Error reading VCF record before opening the file.");
return(false);
}
if(ifeof(filePtr))
{
// End of file, just return false.
return(false);
}
// Read the chromosome.
if(!readTilTab(filePtr, myChrom))
{
if(myChrom.empty())
{
// EOF.
return(false);
}
// Not an empty line.
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record CHROM.");
return(false);
}
// Read the 1-based Position
std::string strPos;
if(!readTilTab(filePtr, strPos))
{
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record POS.");
return(false);
}
else
{
// Read the position, so convert to an integer.
my1BasedPosNum = atoi(strPos.c_str());
}
// Read the ID.
if(!readTilTab(filePtr, myID))
{
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record ID.");
return(false);
}
if(discardRules.discardForID(myID))
{
// Do not keep this id, so consume the rest of the record and
// return the next record.
filePtr->discardLine();
return(read(filePtr, siteOnly, discardRules, sampleSubset));
}
// Read the Ref.
if(!readTilTab(filePtr, myRef))
{
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record REF.");
return(false);
}
// Read the Alt.
myAltArray.clear();
if(!readTilTab(filePtr, myAlt))
{
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record ALT.");
return(false);
}
// Read the Qual.
if(!readTilTab(filePtr, myQual))
{
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record QUAL.");
return(false);
}
else
{
if(myQual != ".")
{
// Read the quality, so convert to an integer.
myQualNum = atof(myQual.c_str());
}
else
{
myQualNum = -1;
}
}
// Read the Filter.
if(!myFilter.read(filePtr))
{
myStatus.setStatus(StatGenStatus::FAIL_PARSE,
"Error reading VCF Record FILTER.");
return(false);
}
// Read the Info (could be the last word in the line or file).
if(!myInfo.read(filePtr))
{
// Found the end of the line after the info field, so return true,
// successfully read the record.
return(true);
}
if(siteOnly)
{
// Do not store genotypes, so just consume the rest of the line.
filePtr->readTilChar("\n");
}
else
{
// Not yet at the end of the line, so read the genotype fields
// (format & samples)
try
{
myGenotype.read(filePtr, sampleSubset);
}
catch(std::exception& e)
{
myDummyString = "Failed parsing the Genotype Fields of " + myChrom + ":" +
std::to_string((long long int)my1BasedPosNum) + " (chr:pos) - " + e.what();
myStatus.setStatus(StatGenStatus::FAIL_PARSE, myDummyString.c_str());
return(false);
}
}
// Found the end of the line, return true since all required fields
// were read.
return(true);
}
bool VcfGenotypeSample::read(IFILE filePtr, VcfGenotypeFormat& format)
{
static const char* GT_DELIM = "\n\t:|/.";
static const int END_GT = 2; // Ends at index 2 or less
static const int PHASED_CHAR_POS = 3;
static const int UNPHASED_CHAR_POS = 4;
static const int MISSING_GT_POS = 5;
// Clear out any previously set values.
reset();
myFormatPtr = &format;
int gtIndex = format.getGTIndex();
// Read the subfields.
SUBFIELD_READ_STATUS readStatus = MORE_SUBFIELDS;
std::string* nextType = NULL;
int subFieldIndex = 0;
while(readStatus == MORE_SUBFIELDS)
{
// Get the field to write into.
if(format.storeIndex(subFieldIndex))
{
nextType = &(myGenotypeSubFields.getNextEmpty());
// Check if this is the GT field.
if(subFieldIndex == gtIndex)
{
// There is a GT field, so set that all GT fields are there.
// if any are missing it will be turned back to false.
myHasAllGenotypeAlleles = true;
// This is the GT field, so parse manually looking to see if it
// is phased and store the genotypes.
int stopChar = END_GT + 1;
// Read until a new subfield is found.
while(stopChar > END_GT)
{
// TODO have an option to autoparse the genotypes?
// todo - store the previous nextType len in order to
// do string conversion to ints...
stopChar = filePtr->readTilChar(GT_DELIM, *nextType);
if(stopChar == PHASED_CHAR_POS)
{
nextType->push_back('|');
myPhased = true;
}
else if(stopChar == UNPHASED_CHAR_POS)
{
nextType->push_back('/');
myUnphased = true;
}
else if(stopChar == MISSING_GT_POS)
{
nextType->push_back('.');
myHasAllGenotypeAlleles = false;
}
}
// Check if this is the END_GT signal.
readStatus = getReadStatus(stopChar);
}
else
{
// more subfields to read.
readStatus = readGenotypeSubField(filePtr, nextType);
}
}
else
{
readStatus = readGenotypeSubField(filePtr, NULL);
}
++subFieldIndex;
}
// subFieldIndex contains the number of fields in this sample.
if(subFieldIndex > format.getOrigNumFields())
{
throw(std::runtime_error("VCF Number of Fields in a Sample does not match the Format."));
}
else if(subFieldIndex < format.getOrigNumFields())
{
// If there are no fields for this sample, enter the missing value.
if(myGenotypeSubFields.size() == 0)
{
myGenotypeSubFields.getNextEmpty() = MISSING_FIELD;
}
}
// Return true if there is a tab - it is just END_OF_FIELD.
return(readStatus == END_OF_FIELD);
}
