void readMTAlleleFreq(const string freqFile, map<int, alleleFrequency> & pos2allelefreq){
// map<int, alleleFrequency> pos2allelefreq;
string line;
igzstream freqAlleleFile;
freqAlleleFile.open(freqFile.c_str());
if (freqAlleleFile.good()){
while ( getline (freqAlleleFile,line)){
vector<string> fields = allTokens(line,'\t');
alleleFrequency freqToadd;
if(fields.size() != 5){
cerr << "line "<<line<<" in file "<<freqFile<<" does not have 5 fields"<<endl;
exit(1);
}
for(int nuc=0;nuc<4;nuc++){
freqToadd.f[nuc]=destringify<double>(fields[nuc+1]);
}
pos2allelefreq[ destringify<int>( fields[0]) ] = freqToadd;
}
freqAlleleFile.close();
}else{
cerr << "Cannot open allele frequency file "<<freqFile<<""<<endl;
exit(1);
}
}
void readMTConsensus(const string consensusFile,
map<int, PHREDgeno> & pos2phredgeno,
int & sizeGenome,
vector<int> & posOfIndels){
string line;
igzstream consensusFD;
consensusFD.open(consensusFile.c_str());
if (consensusFD.good()){
getline (consensusFD,line);
while ( getline (consensusFD,line)){
if (line.empty())
continue;
vector<string> fields = allTokens(line,'\t');
PHREDgeno toadd;
// cerr<<line<<endl;
if(fields.size() != 11){
cerr << "line "<<line<<" in file "<<consensusFile<<" does not have 11 fields"<<endl;
exit(1);
}
if(fields[0][fields[0].size()-1] == 'i'){ //skip insertion
posOfIndels.push_back( destringify<int>( fields[0]) );
continue;
}
if(fields[2] == "D"){ //skip deletions
posOfIndels.push_back( destringify<int>( fields[0]) );
continue;
}
toadd.consensus = fields[2][0];
for(int nuc=0;nuc<4;nuc++){
toadd.phred[nuc] = destringify<double>(fields[nuc+7]);
toadd.perror[nuc] = pow(10.0,toadd.phred[nuc]/(-10.0));
}
pos2phredgeno[ destringify<int>( fields[0]) ] = toadd;
sizeGenome = max( destringify<int>( fields[0]), sizeGenome);
// cout<<destringify<int>( fields[0])<<endl;
}
consensusFD.close();
}else{
cerr << "Cannot open consensus file "<<consensusFile<<""<<endl;
exit(1);
}
}
SimpleVCF::SimpleVCF(string line){
//trimWhiteSpacesBothEnds (&line);
vector<string> fields=allTokens(line,'\t');
corevcf = new CoreVCF(fields);
deleteCore=true;
// cerr<<"Ok "<<endl;
init(fields,corevcf);
// cerr<<"Ok "<<endl;
// cout<<"SimpleVCF "<<this<<"\t"<<deleteCore<<endl;
}
string MistarParser::getHeaderNoDefline(string prefix){
vector<string> fields=allTokens(headerNoDefline,'\n');
vector<string> toreturn;
for(unsigned int i=0;i<fields.size();i++){
if(!fields[i].empty())
toreturn.push_back(prefix+fields[i]);
}
return vectorToString(toreturn,"\n");
}
//IMPLEMENT HEADER
MultiVCFreader::MultiVCFreader(string file,string indexForFile,string chrName,int start,int end,int indelsAhead) {
readAhead=indelsAhead;
rt = new ReadTabix (file,indexForFile,chrName,start,end);
//reading header
istringstream f (rt->getHeader());
string line;
numPop=0;
while (getline(f, line)) {
//std::cout << line << std::endl;
if(strBeginsWith(line,"#CHROM")) {
vector<string> tok = allTokens(line,'\t');
if(tok.size() < 10 ) {
cerr<<"The header line"<<line<<" does not contain enough fields"<<endl;
exit(1);
}
for(unsigned int i=9; i<tok.size(); i++) {
//cerr<<tok[i]<<endl;
numPop++;
populationNames.push_back(tok[i]);
}
}
}
if( numPop == 0 ) {
cerr<<"No populations have been found for file "<<file<<endl;
exit(1);
}
needToPopulateQueue = true;
fullQueue = false;
endQueue = false;
numberOfTimesHasDataWasCalled = 0;
svcfToReturn = 0;
repoCalledHasData = false;
indexInQueueOfIndels = -1;
indexOfLastIndel = 0;
previouslyFoundIndel = false;
tabixMode = true;
textMode = false;
}
void MistarParser::parseHeader(istream & in){
bool firstLine=true;
string line;
while(getline ( in,line)){
//cout<<"line "<<line<<endl;
if(line[0] == '#'){
// cout<<line;
if(firstLine){
if(line != "#MISTAR"){
cerr << "Error: MistarParser first line must be #MISTAR found: " << line <<endl;
exit(1);
}
firstLine=false;
continue;
}
if(strBeginsWith(line, "#chr")){
defline=line;
vector<string> fields=allTokens(line,'\t');
if(fields[0] != "#chr") { cerr<<"Field #1 ("<<fields[0]<<") of header must be #chr in line #"<<line<<"#"<<endl; exit(1); }
if(fields[1] != "coord") { cerr<<"Field #2 of header must be coord "; exit(1); }
if(fields[2] != "REF,ALT"){ cerr<<"Field #3 of header must be REF,ALT "; exit(1); }
if(fields[3] != "root") { cerr<<"Field #4 of header must be root "; exit(1); }
if(fields[4] != "anc") { cerr<<"Field #5 of header must be anc "; exit(1); }
for(unsigned int i=3;i<fields.size();i++){
populationNames->push_back(fields[i]);
numberPopulations++;
}
header+=line+"\n";
break;
}else{
header+=line+"\n";
headerNoDefline+=line+"\n";
}
}else{
cerr << "Error: MistarParser cannot get header" <<endl;
exit(1);
}
}
}
void SimpleVCF::init(const vector<string> & fields, CoreVCF * corevcf_){ //string line){
unresolvedGT=false;
homozygousREF=false;
heterozygous=false;
homozygousALT=false;
indexGenotype=-1;
indexGenotypeQual=-1;
indexDepth=-1;
indexPL=-1;
typeOfData=1;
// fields=allTokens(line,'\t');
// corevcf = corevcf_;
int fieldIndex = corevcf->getFieldIndexAndIncrease();
// cerr<<"fieldIndex "<<fieldIndex<<endl;
//FORMAT FIELDS
rawFormatNames = fields[ corevcf->getFieldIndexINFO()+1 ];
rawFormatValues = fields[fieldIndex];
// cerr<<"rawFormatNames "<<rawFormatNames<<endl;
// cerr<<"rawFormatValues "<<rawFormatValues<<endl;
formatFieldNames = allTokens(rawFormatNames ,':');
formatFieldValues = allTokens(rawFormatValues,':');
if(rawFormatValues == "./."){
unresolvedGT=true;
observedPL=false;
observedGL=false;
haploidCall=false;
}else{
if(formatFieldNames.size() != formatFieldValues.size()){
cerr<<"SimpleVCF: for line "<<vectorToString(fields,"\t")<<" the format field does not have as many fields as the values"<<endl;
exit(1);
}
observedPL=false;
observedGL=false;
haploidCall=false;
for(unsigned int i=0;i<formatFieldNames.size();i++){
// cerr<<"formatFieldNames["<<i<<"] "<<formatFieldNames[i]<<" = "<<formatFieldValues[i]<<endl;
if(formatFieldNames[i] == "GT"){
indexGenotype =i;
formatFieldGT= formatFieldValues[i];
bool determinedGenotype=false;
//Taken from http://www.broadinstitute.org/gatk/guide/topic?name=intro
if(formatFieldGT == "./."){ determinedGenotype=true; unresolvedGT=true; }
if(formatFieldGT == "0"){ determinedGenotype=true; homozygousREF=true; haploidCall=true; }
if(formatFieldGT == "1"){ determinedGenotype=true; homozygousALT=true; haploidCall=true; }
if(formatFieldGT == "0/0"){ determinedGenotype=true; homozygousREF=true; }
if(formatFieldGT == "0|0"){ determinedGenotype=true; homozygousREF=true; }
if(formatFieldGT == "0/1"){ determinedGenotype=true; heterozygous=true; }
if(formatFieldGT == "0|1"){ determinedGenotype=true; heterozygous=true; }
if(formatFieldGT == "1|0"){ determinedGenotype=true; heterozygous=true; }
if(formatFieldGT == "1/1"){ determinedGenotype=true; homozygousALT=true; }
if(formatFieldGT == "1|1"){ determinedGenotype=true; homozygousALT=true; }
if(formatFieldGT == "1/2"){ determinedGenotype=true; heterozygousALT=true; } //has first alt and second alt
if(formatFieldGT == "1|2"){ determinedGenotype=true; heterozygousALT=true; } //has first alt and second alt
if(formatFieldGT == "2/1"){ determinedGenotype=true; heterozygousALT=true; } //has first alt and second alt
if(formatFieldGT == "2|1"){ determinedGenotype=true; heterozygousALT=true; } //has first alt and second alt
if(formatFieldGT == "0|2"){ determinedGenotype=true; heterozygous2ndALT=true; } //has ref and second alt
if(formatFieldGT == "0/2"){ determinedGenotype=true; heterozygous2ndALT=true; } //has ref and second alt
if(formatFieldGT == "2/0"){ determinedGenotype=true; heterozygous2ndALT=true; } //has ref and second alt
if(formatFieldGT == "2|0"){ determinedGenotype=true; heterozygous2ndALT=true; } //has ref and second alt
if(formatFieldGT == "2/2"){ determinedGenotype=true; homozygous2ndALT=true; } //twice the second alt
if(formatFieldGT == "2|2"){ determinedGenotype=true; homozygous2ndALT=true; } //twice the second alt
//for more than 3
if(!determinedGenotype){
vector<string> fieldsOfGT = allTokens(formatFieldGT ,'/');
if(fieldsOfGT.size() == 2){
// int alleleCFirst = destringify<int> (fieldsOfGT[0]);
// int alleleC2nd = destringify<int> (fieldsOfGT[1]);
if(isPositiveInt(fieldsOfGT[0]) &&
isPositiveInt(fieldsOfGT[1]) ){
determinedGenotype=true; unresolvedGT=true;
}
}else{
vector<string> fieldsOfGT = allTokens(formatFieldGT ,'|');
if(fieldsOfGT.size() == 2){
// int alleleCFirst = destringify<int> (fieldsOfGT[0]);
// int alleleC2nd = destringify<int> (fieldsOfGT[1]);
if(isPositiveInt(fieldsOfGT[0]) &&
isPositiveInt(fieldsOfGT[1]) ){
determinedGenotype=true; unresolvedGT=true;
}
}else{
}
}
}
// if(formatFieldGT == "0/3" ||
// formatFieldGT == "3/3" ||
// formatFieldGT == "3/3" ||
// ){ determinedGenotype=true; unresolvedGT=true;
if(!determinedGenotype){
cerr<<"SimpleVCF: unable to determine genotype for line "<<vectorToString(fields,"\t")<<" field=#"<<formatFieldGT<<"#"<<endl;
exit(1);
}
continue;
}
if(formatFieldNames[i] == "GQ"){
if(formatFieldValues[i] == "."){
indexGenotypeQual =i;
formatFieldGQ=0.0;
}else{
indexGenotypeQual =i;
formatFieldGQ=destringify<float>(formatFieldValues[i]);
}
continue; }
if(formatFieldNames[i] == "DP"){ indexDepth =i; formatFieldDP=destringify<int> (formatFieldValues[i]); continue;}
if(formatFieldNames[i] == "GL"){
observedGL=true;
if(observedPL){
cerr<<"SimpleVCF: cannot observed both GL and PL "<<vectorToString(fields,"\t")<<""<<endl;
exit(1);
}
indexPL = i;
formatFieldGL = formatFieldValues[i];
vector<string> glfields = allTokens(formatFieldGL,',');
if(glfields.size() == 2){ //haploid calls (e.g. X for a male)
if(!haploidCall){
cerr<<"SimpleVCF: cannot observed 2 GL fields for a non-haploid record "<<vectorToString(fields,"\t")<<""<<endl;
exit(1);
}
formatFieldPLHomoRef = int(-10.0*destringify<double>(glfields[0]));
formatFieldPLHetero = -1000000; //very unlikely
formatFieldPLHomoAlt = int(-10.0*destringify<double>(glfields[1]));
}else{
if(glfields.size() == 3){ //biallelic
formatFieldPLHomoRef = int(-10.0*destringify<double>(glfields[0]));
formatFieldPLHetero = int(-10.0*destringify<double>(glfields[1]));
formatFieldPLHomoAlt = int(-10.0*destringify<double>(glfields[2]));
}else{
if(glfields.size() == 6){ //triallelic
//according to VCF docs it has the following order AA,AB,BB,AC,BC,CC
formatFieldPLHomoRef = int(-10.0*destringify<double>(glfields[0])); //r-r
formatFieldPLHetero1 = int(-10.0*destringify<double>(glfields[1])); //r-a1
formatFieldPLHomoAlt1 = int(-10.0*destringify<double>(glfields[2])); //a1-a1
formatFieldPLHetero2 = int(-10.0*destringify<double>(glfields[3])); //r-a2
formatFieldPLHetero12 = int(-10.0*destringify<double>(glfields[4])); //a1-a2
formatFieldPLHomoAlt2 = int(-10.0*destringify<double>(glfields[5])); //a2-a2
}else{
cerr<<"SimpleVCF: for line "<<vectorToString(fields,"\t")<<" the GL field does not have 3 or 6 fields"<<endl;
exit(1);
}
}
}
}
if(formatFieldNames[i] == "PL"){
observedPL=true;
if(observedGL){
cerr<<"SimpleVCF: cannot observed both GL and PL "<<vectorToString(fields,"\t")<<""<<endl;
exit(1);
}
indexPL = i;
if(formatFieldValues[i] == "."){
formatFieldPL = formatFieldValues[i];
unresolvedGT=true;
continue;
}
formatFieldPL = formatFieldValues[i];
vector<string> plfields = allTokens(formatFieldPL,',');
if(plfields.size() == 3){ //biallelic
formatFieldPLHomoRef = destringify<int>(plfields[0]);
formatFieldPLHetero = destringify<int>(plfields[1]);
formatFieldPLHomoAlt = destringify<int>(plfields[2]);
}else{
if(plfields.size() == 6){ //triallelic
//according to VCF docs it has the following order AA,AB,BB,AC,BC,CC
formatFieldPLHomoRef = destringify<int>(plfields[0]); //r-r
formatFieldPLHetero1 = destringify<int>(plfields[1]); //r-a1
formatFieldPLHomoAlt1 = destringify<int>(plfields[2]); //a1-a1
formatFieldPLHetero2 = destringify<int>(plfields[3]); //r-a2
formatFieldPLHetero12 = destringify<int>(plfields[4]); //a1-a2
formatFieldPLHomoAlt2 = destringify<int>(plfields[5]); //a2-a2
}else{
cerr<<"SimpleVCF: for line "<<vectorToString(fields,"\t")<<" the PL field does not have 3 or 6 fields"<<endl;
exit(1);
}
}
continue;
}
//To uncomment the fields to get these fields
if(formatFieldNames[i] == "A"){
vector<string> adfield = allTokens( formatFieldValues[i] ,',');
for(unsigned int j=0;j<adfield.size();j++){
countA.push_back( destringify<int>( adfield[j]) );
}
continue;
}
if(formatFieldNames[i] == "C"){
vector<string> adfield = allTokens( formatFieldValues[i] ,',');
for(unsigned int j=0;j<adfield.size();j++){
countC.push_back( destringify<int>( adfield[j]) );
}
continue;
}
if(formatFieldNames[i] == "G"){
vector<string> adfield = allTokens( formatFieldValues[i] ,',');
for(unsigned int j=0;j<adfield.size();j++){
countG.push_back( destringify<int>( adfield[j]) );
}
continue;
}
if(formatFieldNames[i] == "T"){
vector<string> adfield = allTokens( formatFieldValues[i] ,',');
for(unsigned int j=0;j<adfield.size();j++){
countT.push_back( destringify<int>( adfield[j]) );
}
continue;
}
}
}
// cout<<getADforA()<<endl;
// cout<<getADforC()<<endl;
// cout<<getADforG()<<endl;
// cout<<getADforT()<<endl;
// cerr<<"end"<<endl;
}
indexData intern_readIndex(string filename){
string line;
ifstream myFile;
indexData toReturn;
toReturn.mlindex1=0;
toReturn.mlindex2=0;
bool isFirstLine =true;
//initialize the values for the likelihood of matches or mismatches
for(int i=0;i<64;i++){
if(i == 0)
likeMatch[i] = -3.0; // this is vrong, hope it's never accessed
else
likeMatch[i] = log1p( -pow(10.0,i/-10.0) )/log(10);
likeMismatch[i] = i/-10.0;
#ifdef DEBUG2
cout<<"qual = "<<i<<endl;
cout<<likeMatch[i]<<endl;
cout<<likeMismatch[i]<<endl;
#endif
}
//reading the files
// myFile.open(filename.c_str(), ios::in);
// if (myFile.is_open()){
vector<string> allLinesIndex = allTokens(filename,'\n');
//while ( getline (myFile,line)){
for(unsigned int i=0;i<allLinesIndex.size();i++){
line = allLinesIndex[i];
if(line.empty())
continue;
line+=' ';
// cerr<<"line #"<<line<<"#"<<toReturn.isDoubleIndex<<endl;
if(isFirstLine){
if(line[0] == '#'){
unsigned int i=0;
int numberOfFields=0;
bool inWS=true;
while(i<line.length()){
if( isspace(line[i])){
inWS=true;
}else{
if(inWS){
numberOfFields++;
}
inWS=false;
}
i++;
}
if(numberOfFields==2){
toReturn.isDoubleIndex=false;
}else{
if(numberOfFields==3 || numberOfFields==5){
toReturn.isDoubleIndex=true;
}else{
cerr << "Must have 2, 3 or 5 fields"<<endl;
exit(1);
}
}
}else{
cerr << "First line must begin with #"<<endl;
exit(1);
}
isFirstLine=false;
}else{
int i=0;
int fieldIndex=0;
bool inWS=false;
int lastOneNW=0;
string foundName;
while(i<int(line.length())){
if( isspace(line[i]) && i==0){
cerr<<line<<endl;
cerr << "First character cannot be a space"<<endl;
exit(1);
}
if( isspace(line[i]) ){
if(!inWS){ //found a field
//first field, first index
if(fieldIndex==0){
toReturn.indices1.push_back(toUpperCase(line.substr(lastOneNW,i-lastOneNW)));
if(toReturn.mlindex1 < (i-lastOneNW)){
toReturn.mlindex1 =(i-lastOneNW);
}
}else{
//second field, either name of single ind or second index
if(fieldIndex==1){
if(toReturn.isDoubleIndex){
toReturn.indices2.push_back(toUpperCase(line.substr(lastOneNW,i-lastOneNW)));
if(toReturn.mlindex2 < (i-lastOneNW)){
toReturn.mlindex2 =(i-lastOneNW);
}
}else{
foundName=line.substr(lastOneNW,i-lastOneNW);
//duplicated names ?
if(toReturn.namesMap.find( foundName ) != toReturn.namesMap.end()){
cerr<<"Warning: The sequence name is duplicated "<<foundName<<endl;
//exit(1);
}else{
toReturn.namesMap[ foundName ] = "";
}
toReturn.names.push_back( foundName );
}
}else if(fieldIndex==2){
//sequence name when two indices
if(toReturn.isDoubleIndex){
//duplicated names
foundName=line.substr(lastOneNW,i-lastOneNW);
if(toReturn.namesMap.find( foundName ) != toReturn.namesMap.end()){
cerr<<"Warning: The sequence name is duplicated "<<foundName<<endl;
//exit(1);
}else{
toReturn.namesMap[ foundName ] = "";
}
toReturn.names.push_back( foundName );
}else{
//it's a comment for single index
toReturn.namesMap[ foundName ] += line.substr(lastOneNW,i-lastOneNW);
// cerr<<"Single index file cannot have 3 fields"<<endl;
// exit(1);
}
}else{
//it's a comment again
toReturn.namesMap[ foundName ] += line.substr(lastOneNW,i-lastOneNW);
}
}
fieldIndex++;
}
inWS=true;
}else{
if(inWS)
lastOneNW=i;
inWS=false;
}
i++;
} //ending while(i<line.length()){
} // ending else firstline
} // ending while myFile.good() ){
//checking for size
// cout<<toReturn.indices1.size()<<endl;
// cout<<toReturn.indices2.size()<<endl;
// cout<<toReturn.names.size()<<endl;
if(toReturn.isDoubleIndex)
if((toReturn.indices1.size() != toReturn.indices2.size()) ){
cerr << "Size of the fields inconsistent "<<filename<<endl;
exit(1);
}
if(toReturn.indices1.size() != toReturn.names.size() ){
cerr << "Size of the fields inconsistent "<<filename<<endl;
exit(1);
}
//checking for valid dna
for(unsigned int i=0;i<toReturn.indices1.size();i++){
if(!isValidDNA(toReturn.indices1[i])){
cerr << "Index " << toReturn.indices1[i] <<" is not a valid DNA sequence"<<endl;
exit(1);
}
if(toReturn.isDoubleIndex)
if(!isValidDNA(toReturn.indices2[i])){
cerr << "Index " << toReturn.indices2[i] <<" is not a valid DNA sequence"<<endl;
exit(1);
}
}
return toReturn;
}
MultiVCFreader::MultiVCFreader(string file,int indelsAhead) {
readAhead=indelsAhead;
numberOfTimesHasDataWasCalled=0;
svcfToReturn=0;
vcfFile.open(file.c_str(), ios::in); // open the streams
if (vcfFile.good()) {
//fine
} else {
cerr<<"Unable to open the file "<<file<<endl;
exit(1);
}
bool firstLine=true;
bool haveCaptureCHROM=false;
numPop=0;
while(1) {
bool flag=getline(vcfFile,currentline);
if(!flag) {
cerr<<"ERROR file : "+file+" is probably empty"<<endl;
exit(1);
}
if(firstLine) {
if(currentline.length() > 0 && currentline[0] != '#') {
cerr<<"ERROR first line in "<<file<<"does not begin with #"<<endl;
exit(1);
}
firstLine = false;
}
if(!firstLine) {
if(currentline.length() > 0) {
if(currentline[0] == '#') {
if(strBeginsWith(currentline,"#CHROM")) {
haveCaptureCHROM=true;
vector<string> tok = allTokens(currentline,'\t');
if(tok.size() < 10 ) {
cerr<<"The header line"<<currentline<<" does not contain enough fields for file "<<file<<endl;
exit(1);
}
for(unsigned int i=9; i<tok.size(); i++) {
//cerr<<tok[i]<<endl;
numPop++;
populationNames.push_back(tok[i]);
}
break;
}
// cerr<<"ERROR first line in "<<file<<"does not begin with #"<<endl;
// return 1;
} else {
break;
}
}
}
}//end while(1)
// vcfFile.close();
// //vcfFile.seekg(0, std::ios::beg);
// vcfFile.open(file.c_str(), ios::in); // open the streams
// if (vcfFile.good()) {
// //fine
// }else{
// cerr<<"Unable to open the file for second pass "<<file<<endl;
// exit(1);
// }
if( numPop == 0 ) {
cerr<<"No populations have been found for file "<<file<<endl;
exit(1);
}
if(!haveCaptureCHROM) {
cerr<<"The header with #CHROM has not been found in file:"<<file<<endl;
exit(1);
}
needToPopulateQueue = true;
fullQueue = false;
endQueue = false;
repoCalledHasData = false;
indexInQueueOfIndels=-1;
indexOfLastIndel=0;
previouslyFoundIndel=false;
tabixMode = false;
textMode = true;
}
bool MultiVCFreader::hasData() {
if(repoCalledHasData) {
repoCalledHasData=false;
return true;
}
numberOfTimesHasDataWasCalled++;
//if first call and queue empty, populate
if(needToPopulateQueue) {
//cout<<"hasData()"<<endl;
bool loop=true;
int indexQueue=0;
while(loop) {
if(getNextLine()) {
#ifdef DEBUG
cout<<"currentline "<<currentline<<endl;
#endif
vector<SimpleVCF *> * svcfvec = new vector<SimpleVCF *>();
vector<string> fieldTab = allTokens(currentline,'\t');
CoreVCF * corevcf = new CoreVCF(fieldTab);
for(int k=0; k<numPop; k++) {
SimpleVCF * svcf = new SimpleVCF (fieldTab,corevcf,k==0);
svcfvec->push_back(svcf);
}
//SimpleVCF * svcfvec = new SimpleVCF(currentline);
#ifdef DEBUG
//cout<<"new1 "<<svcf<<endl;
#endif
if(queueOfVCFs.size() != 0 ) {
flagCpG( queueOfVCFs.back()->at(0) , svcfvec->at(0) );
}
// cout<<"Adding "<<*svcf<<endl;
queueOfVCFs.push_back(svcfvec);
if(svcfvec->at(0)->containsIndel()) {
if(indexInQueueOfIndels == -1)
indexInQueueOfIndels=indexQueue;
}
indexQueue++;
if(queueOfVCFs.size() == (readAhead+1)) {
loop=false;
}
} else {
loop=false;
}
}
if(queueOfVCFs.size() == (readAhead+1)) { //+1 for CPGs
fullQueue=true;
} else {
endQueue=true;
}
needToPopulateQueue=false;
}
//if subsequent call, and queue full
if(fullQueue) {
if(getNextLine()) {
// SimpleVCF * svcf = new SimpleVCF(currentline);
vector<SimpleVCF *> * svcfvec = new vector<SimpleVCF *>();
vector<string> fieldTab = allTokens(currentline,'\t');
CoreVCF * corevcf = new CoreVCF(fieldTab);
for(int k=0; k<numPop; k++) {
SimpleVCF * svcf = new SimpleVCF (fieldTab,corevcf,k==0);
svcfvec->push_back(svcf);
}
#ifdef DEBUG
//cout<<"new2 "<<*svcf<<endl;
#endif
// cout<<"size "<<queueOfVCFs.size()<<endl;
if(queueOfVCFs.size() != 0 ) {
flagCpG( queueOfVCFs.back()->at(0),svcfvec->at(0));
}
queueOfVCFs.push_back(svcfvec);
if(svcfvec->at(0)->containsIndel()) {
if(indexInQueueOfIndels == -1)
indexInQueueOfIndels=queueOfVCFs.size()-1;
}
} else {
fullQueue=false;
endQueue=true;
}
}
//if final calls and queue not max size
if(endQueue) {
//nothing to do
}
bool stillHasData=!(queueOfVCFs.empty());
// if(!stillHasData){ //getData() should not get called in this case, hence no deallocation
// cout<<"delete1 "<<svcfToReturn<<endl;
// delete svcfToReturn;
// }
return stillHasData;
}
void readNucSubstitionFreq(const string filename,vector<probSubstition> & subVec){
igzstream subFP;
subFP.open(filename.c_str(), ios::in);
// unsigned int counterCont=0;
if (subFP.good()){
vector<string> fields;
string line;
//header
if ( !getline (subFP,line)){
cerr << "Unable to open file "<<filename<<endl;
exit(1);
}
fields = allTokens(line,'\t');
if(fields.size() != 12){
cerr << "line from error profile does not have 12 fields "<<line<<endl;
exit(1);
}
//probs
while ( getline (subFP,line)){
fields = allTokens(line,'\t');
if(fields.size() != 12){
cerr << "line from error profile does not have 12 fields "<<line<<endl;
exit(1);
}
substitutionRates tempFreq;
probSubstition toaddSub;
for(unsigned int k=0;k<=9;k+=3){
for(unsigned int t=0;t<=2;t++){
tempFreq.s[k+t]=destringify<double>(fields[k+t]);
}
}
int indexFirstArray =0;
int indexSecondArray=0;
for(int nuc1=0;nuc1<4;nuc1++){
double sumMismatchProb=0.0;
int indexInArrayMatch=1;
for(int nuc2=0;nuc2<4;nuc2++){
if(nuc1==nuc2){ // prob of error is 0 if both nucleotides are identical
indexInArrayMatch = indexFirstArray;
toaddSub.s[indexFirstArray++] = 1.0;
}else{ // rely on the substitution frequency
sumMismatchProb += tempFreq.s[indexSecondArray];
toaddSub.s[indexFirstArray++] = tempFreq.s[indexSecondArray++];
}
}
toaddSub.s[indexInArrayMatch] = 1.0 - sumMismatchProb;
}
// for(int nuc1=0;nuc1<4;nuc1++){
// for(int nuc2=0;nuc2<4;nuc2++){
// cout<<(nuc1*4+nuc2)<<"\t"<<toaddSub.s[nuc1*4+nuc2]<<endl;
// }
// }
// exit(1);
subVec.push_back( toaddSub );
}
subFP.close();
}else{
cerr << "Unable to open file "<<filename<<endl;
exit(1);
}
}
void readIlluminaError(const string errFile,probSubstition & illuminaErrorsProb){
igzstream errFileSt;
errFileSt.open(errFile.c_str(), ios::in);
// unsigned int counterCont=0;
if (errFileSt.good()){
vector<string> fields;
string line;
//header
if ( !getline (errFileSt,line)){
cerr << "Unable to open file "<<errFile<<endl;
exit(1);
}
fields = allTokens(line,'\t');
if(fields.size() != 12){
cerr << "line from error profile does not have 12 fields "<<line<<endl;
exit(1);
}
//raw sums
if ( !getline (errFileSt,line)){
cerr << "Unable to open file "<<errFile<<endl;
exit(1);
}
fields = allTokens(line,'\t');
if(fields.size() != 12){
cerr << "line from error profile does not have 12 fields "<<line<<endl;
exit(1);
}
//probs
if ( !getline (errFileSt,line)){
cerr << "Unable to open file "<<errFile<<endl;
exit(1);
}
fields = allTokens(line,'\t');
if(fields.size() != 12){
cerr << "line from error profile does not have 12 fields "<<line<<endl;
exit(1);
}
substitutionRates tempFreq;
for(unsigned int k=0;k<=9;k+=3){
for(unsigned int t=0;t<=2;t++){
tempFreq.s[k+t]=destringify<double>(fields[k+t]);
//cerr<<freqIlluminaError.s[k+t]<<endl;
}
}
int indexFirstArray =0;
int indexSecondArray=0;
for(int nuc1=0;nuc1<4;nuc1++){
for(int nuc2=0;nuc2<4;nuc2++){
if(nuc1==nuc2) // prob of error is 0 if both nucleotides are identical
illuminaErrorsProb.s[indexFirstArray++]=0.0;
else // rely on the substitution frequency
illuminaErrorsProb.s[indexFirstArray++]=tempFreq.s[indexSecondArray++];
}
}
errFileSt.close();
}else{
cerr << "Unable to open file "<<errFile<<endl;
exit(1);
}
}
bool MistarParser::hasData(){
if(numberOfTimesHasDataWasCalled!=-1){
// cout<<"delete"<<endl;
//cerr<<"del "<<allRecToReturn<<endl;
//delete(allRecToReturn->vectorAlleles);
delete(allRecToReturn);
numberdel++;
}else{
numberOfTimesHasDataWasCalled=0;
}
numberOfTimesHasDataWasCalled++;
// string line;
//if(getline ( *myFilezipped,line)){
if(getNextLine()){
numbernew++;
allRecToReturn = new AlleleRecords();
// cerr<<"new "<<allRecToReturn<<endl;
//allRecToReturn->vectorAlleles = new vector<SingleAllele>();
// cout<<"currentline "<<currentline<<endl;
vector<string> fields=allTokens(currentline,'\t');
if(fields.size() != (numberPopulations+3)){
cerr << "Error: MistarParser the following line should have "<<(numberPopulations+3)<<" fields " << currentline <<endl;
exit(1);
}
if(fields[2].length() != 3){
cerr << "Error: MistarParser the following line " << currentline <<" does not have 2 comma separated alleles"<<endl;
exit(1);
}
allRecToReturn->chr = fields[0];
allRecToReturn->coordinate = destringify<unsigned int>(fields[1]);
allRecToReturn->ref = fields[2][0];
allRecToReturn->alt = fields[2][2];
if(allRecToReturn->ref == allRecToReturn->alt){
cerr << "Error: MistarParser the following line " << currentline <<" the reference is equal to the alt allele, exiting"<<endl;
exit(1);
}
allRecToReturn->vectorAlleles = new vector<SingleAllele>();
for(unsigned int i=3;i<fields.size();i++){
unsigned int indexComma=0;
unsigned int indexColon=0;
for(unsigned int k=0;k<fields[i].size();k++){
if(fields[i][k]==',')
indexComma=k;
if(fields[i][k]==':')
indexColon=k;
}
if(indexComma == 0 || indexColon == 0 ){
cerr << "Error: MistarParser problem with the following line " << currentline <<" cannot get allele count"<<endl;
exit(1);
}
SingleAllele sa (destringify<int>( fields[i].substr(0,indexComma)),
destringify<int>( fields[i].substr(indexComma+1,indexColon)),
destringify<bool>(fields[i].substr(indexColon+1)) );
allRecToReturn->vectorAlleles->push_back(sa);
}
if( allRecToReturn->vectorAlleles->size() != numberPopulations){
cerr << "Error: MistarParser problem with the following line " << currentline <<" number of allele count read is not "<<numberPopulations<<endl;
exit(1);
}
return true;
}else{//if has no data
if(textMode){
myFilezipped->close();
}
// else
// myFile->close();
return false;
}
return false;
}
