// specific to upgrade from VCF 3.3 to VCF 4.0
void VcfFile::upgradeMetaLines() {
std::vector<int> toDelete;
for(int i=0; i < asMetaKeys.Length(); ++i) {
if ( asMetaKeys[i].Compare("fileformat") == 0 ) {
if ( asMetaValues[i].Compare("VCFv3.3") == 0 ) {
asMetaValues[i] = "VCFv4.0";
}
}
else if ( asMetaKeys[i].Compare("FILTER") == 0 ) {
toDelete.push_back(i);
}
else if ( asMetaKeys[i].Compare("FORMAT") == 0 ) {
StringArray tok;
tok.ReplaceColumns(asMetaValues[i],',');
asMetaValues[i].printf("<ID=%s,Number=%s,Type=%s,Description=%s>",tok[0].c_str(),tok[2].c_str(),tok[1].c_str(),tok[3].c_str());
}
}
for(int i=0; i < (int)toDelete.size(); ++i) {
asMetaValues.Delete(toDelete[toDelete.size()-i-1]);
asMetaKeys.Delete(toDelete[toDelete.size()-i-1]);
}
// add INFO and FILTER?
}
bool FilterStat::appendStatVcf(const char* file) {
VcfFile vcf;
vcf.setSiteOnly(false);
vcf.setParseValues(true);
vcf.setParseGenotypes(false);
vcf.setParseDosages(false);
vcf.openForRead(file,1);
VcfMarker* pMarker;
StringArray tok;
for( int i=0, j=0; vcf.iterateMarker(); ++i, ++j ) {
pMarker = vcf.getLastMarker();
if ( sChrom.Compare(pMarker->sChrom) != 0 ) {
Logger::gLogger->error("Chromosome name does not match - %s vs %s",sChrom.c_str(),pMarker->sChrom.c_str());
}
while ( vPos[j] < pMarker->nPos ) { ++j; }
if ( vPos[j] > pMarker->nPos ) {
Logger::gLogger->error("Position %s:%d is not observed in the anchor VCF",sChrom.c_str(),pMarker->nPos);
}
std::vector<int> vAlleles;
std::vector<int> vStrands;
//fprintf(stderr,"%s:%d\t%s\n",pMarker->sChrom.c_str(),pMarker->nPos,pMarker->asFormatKeys[0].c_str());
for(int k=0; k < pMarker->asFormatKeys.Length(); ++k) {
if ( pMarker->asFormatKeys[k].Compare("BASE") == 0 ) {
tok.ReplaceColumns(pMarker->asSampleValues[k],',');
for(int l=0; l < tok.Length(); ++l) {
if ( tok[l].Compare(vAl1[j].c_str()) == 0 ) {
vAlleles.push_back(0);
}
else if ( tok[l].Compare(vAl2[j].c_str()) == 0 ) {
vAlleles.push_back(1);
}
else {
vAlleles.push_back(2);
}
}
}
else if ( pMarker->asFormatKeys[k].Compare("STRAND") == 0 ) {
tok.ReplaceColumns(pMarker->asSampleValues[k],',');
for(int l=0; l < tok.Length(); ++l) {
if ( tok[l].Compare("F") == 0 ) {
vStrands.push_back(0);
}
else {
vStrands.push_back(1);
}
}
}
}
//fprintf(stderr,"%s:%d\t%d",pMarker->sChrom.c_str(),pMarker->nPos,(int)vAlleles.size());
//for(int k=0; k < (int) vAlleles.size(); ++k) {
// fprintf(stderr,"\t%d",vAlleles[k]*2+vStrands[k]);
//}
//fprintf(stderr,"\n");
if ( vAlleles.size() != vStrands.size() ) {
Logger::gLogger->error("Alleles and Strands do not match in size at %s:%d, in %s",pMarker->sChrom.c_str(), pMarker->nPos, file);
}
// updates the counts - needs synchronization
{
//boost::mutex::scoped_lock lock(mutex);
for(int k=0; k < (int) vAlleles.size(); ++k) {
++(vCounts[FILTER_STAT_COUNTS*j + vAlleles[k]*2 + vStrands[k]]);
}
}
}
return true;
}
int VcfMac::execute(int argc, char **argv)
{
String inputVcf = "";
int minAC = -1;
String sampleSubset = "";
String filterList = "";
bool params = false;
IntervalTree<int> regions;
std::vector<int> intersection;
// Read in the parameters.
ParameterList inputParameters;
BEGIN_LONG_PARAMETERS(longParameterList)
LONG_PARAMETER_GROUP("Required Parameters")
LONG_STRINGPARAMETER("in", &inputVcf)
LONG_PARAMETER_GROUP("Optional Parameters")
LONG_STRINGPARAMETER("sampleSubset", &sampleSubset)
LONG_INTPARAMETER("minAC", &minAC)
LONG_STRINGPARAMETER("filterList", &filterList)
LONG_PARAMETER("params", ¶ms)
LONG_PHONEHOME(VERSION)
END_LONG_PARAMETERS();
inputParameters.Add(new LongParameters ("Input Parameters",
longParameterList));
inputParameters.Read(argc-1, &(argv[1]));
// Check that all files were specified.
if(inputVcf == "")
{
usage();
inputParameters.Status();
std::cerr << "Missing \"--in\", a required parameter.\n\n";
return(-1);
}
if(params)
{
inputParameters.Status();
}
// Open the two input files.
VcfFileReader inFile;
VcfHeader header;
VcfRecord record;
// Open the file
if(sampleSubset.IsEmpty())
{
inFile.open(inputVcf, header);
}
else
{
inFile.open(inputVcf, header, sampleSubset, NULL, NULL);
}
// Add the discard rule for minor allele count.
if(minAC >= 0)
{
inFile.addDiscardMinMinorAlleleCount(minAC, NULL);
}
if(!filterList.IsEmpty())
{
// Open the filter list.
IFILE regionFile = ifopen(filterList, "r");
String regionLine;
StringArray regionColumn;
int start;
int end;
int intervalVal = 1;
if(regionFile == NULL)
{
std::cerr << "Failed to open " << filterList
<< ", so keeping all positions\n";
filterList.Clear();
}
else
{
while( regionFile->isOpen() && !regionFile->ifeof())
{
// Read the next interval
regionLine.Clear();
regionLine.ReadLine(regionFile);
if(regionLine.IsEmpty())
{
// Nothing on this line, continue to the next.
continue;
}
regionColumn.ReplaceColumns(regionLine, ' ');
if(regionColumn.Length() != 2)
{
std::cerr << "Improperly formatted region line: "
<< regionLine << "; skipping to the next line.\n";
continue;
}
// Convert the columns to integers.
if(!regionColumn[0].AsInteger(start))
{
// The start position (1st column) is not an integer.
std::cerr << "Improperly formatted region line, start position "
<< "(1st column) is not an integer: "
<< regionColumn[0]
<< "; Skipping to the next line.\n";
continue;
}
if(!regionColumn[1].AsInteger(end))
{
// The start position (1st column) is not an integer.
std::cerr << "Improperly formatted region line, end position "
<< "(2nd column) is not an integer: "
<< regionColumn[1]
<< "; Skipping to the next line.\n";
continue;
}
// Add 1-based inclusive intervals.
regions.add(start,end, intervalVal);
}
}
}
int numReadRecords = 0;
while( inFile.readRecord(record))
{
if(!filterList.IsEmpty())
{
// Check if the region should be kept.
intersection.clear();
regions.get_intersecting_intervals(record.get1BasedPosition(), intersection);
if(intersection.empty())
{
// not in the interval, so continue to the next record.
continue;
}
}
++numReadRecords;
// Loop through the number of possible alternates.
unsigned int numAlts = record.getNumAlts();
int minAlleleCount = -1;
int curAlleleCount = 0;
int totalAlleleCount = 0;
for(unsigned int i = 0; i <= numAlts; i++)
{
curAlleleCount = record.getAlleleCount(i);
if((minAlleleCount == -1) ||
(curAlleleCount < minAlleleCount))
{
minAlleleCount = curAlleleCount;
}
totalAlleleCount += curAlleleCount;
}
if(totalAlleleCount != 0)
{
double maf = (double)minAlleleCount/totalAlleleCount;
std::cout << record.getIDStr()
<< "\t" << minAlleleCount
<< "\t" << maf << "\n";
}
}
inFile.close();
// std::cerr << "\n\t# Records: " << numReadRecords << "\n";
// return success.
return(0);
}
