void GenomeRegionSeqStats::OutputClusters(String &outFile, int minDepth, double minAvgDepth, int minClusterSize)
{
FILE *fh = fopen(outFile.c_str(), "w");
if(fh==NULL) error("Open cluster output file %s failed!\n", outFile.c_str());
bool inCluster = false;
int winSize = 0;
double avgDepth = 0;
int totalDepth = 0;
uint32_t cstart, cend; //cluster start and cluster end
for(uint32_t i=0; i<depth.size(); i++)
{
if(depth[i]>=minDepth)
{
if(inCluster==false)
{
inCluster = true;
cstart = i;
}
winSize++;
totalDepth += depth[i];
}
else
{
if(inCluster==false) continue;
//std::cout<<totalDepth<<" "<<winSize<<" "<<inCluster<<":"<<minClusterSize<<":"<<avgDepth<<":"<<minAvgDepth<<std::endl;
avgDepth = double(totalDepth)/winSize;
if(winSize<minClusterSize || avgDepth<minAvgDepth)
{
winSize = 0;
totalDepth = 0;
inCluster = false;
continue;
}
winSize = 0;
totalDepth = 0;
inCluster = false;
cend = i-1;
int atCnt, gcCnt;
double gc = CalcRegionGCContent(referencegenome, cstart, cend, atCnt, gcCnt);
String chr; int start, end;
referencegenome.getChromosomeAndIndex(chr, cstart);
StringArray tokens;
tokens.ReplaceTokens(chr, ":");
chr = tokens[0];
start = tokens[1].AsInteger();
end = start + (cend-cstart);
fprintf(fh, "%s\t%u\t%u\t%.2f\t%.2f\n", chr.c_str(), start, end, avgDepth, gc);
}
}
fclose(fh);
}
void GenomeRegionSeqStats::LoadRegionList(String &inputList)
{
FILE *in = fopen(inputList.c_str(), "r");
if(in==NULL) error("Open region input file %s failed!\n", inputList.c_str());
StringArray tokens;
String buffer;
while(!feof(in))
{
buffer.ReadLine(in);
if (buffer.IsEmpty() || buffer[0] == '#') continue;
tokens.ReplaceTokens(buffer);
if(tokens.Length()<3)
error("Too few columns: %s\n", buffer.c_str());
String CSE = tokens[0]+":"+tokens[1]+":"+tokens[2];
std::pair<int, int> start_end;
start_end.first = tokens[1].AsInteger();
start_end.second = tokens[2].AsInteger();
if(start_end.first>=start_end.second) // positions are 0-based. Otherwise == is valid
error("Region end is equal or smaller than the start: %s!\n", buffer.c_str());
genomeRegions_lines[tokens[0]].push_back(buffer);
genomeRegions[tokens[0]].push_back(start_end);
genomeRegions_currentIndex[tokens[0]] = 0;
if(tokens.Length()>3) {
groupStats[tokens[3]].segCount++;
groupStats[tokens[3]].totalLen += (start_end.second - start_end.first);
genomeRegionGroups[CSE].push_back(tokens[3]);
}
}
fclose(in);
// Chromosome info
contigs.clear();
std::map<String, vector<std::pair<int, int> > >::iterator p;
for(p=genomeRegions.begin(); p!=genomeRegions.end(); p++)
{
contigs.push_back(p->first);
for(unsigned int i=1; i<genomeRegions[p->first].size(); i++)
if(genomeRegions[p->first][i].first<genomeRegions[p->first][i-1].first)
error("Input coordinates are not in order: %s %d %d!\n", p->first.c_str(),genomeRegions[p->first][i].first,genomeRegions[p->first][i].second);
}
// Group info such as gene names
groups.clear();
std::map<String, Stats>::iterator p2;
for(p2=groupStats.begin(); p2!=groupStats.end(); p2++)
groups.push_back(p2->first);
}
void VerifyBamID::loadSubsetInds(const char* subsetFile) {
if ( ( pPile == NULL ) && ( pGenotypes == NULL ) ) {
if ( subsetInds.size() > 0 ) {
Logger::gLogger->error("VerifyBamID::loadSubsetInds() called multiple times");
}
IFILE f = ifopen(subsetFile,"rb");
String line;
StringArray tok;
while( line.ReadLine(f) > 0 ) {
tok.ReplaceTokens(line,"\t \n\r");
subsetInds.push_back(tok[0].c_str());
}
}
else {
Logger::gLogger->error("VerifyBamID::loadSubsetInds() called after VerifyBamID::loadFiles()");
}
}
bool MarkovParameters::ReadCrossoverRates(const char * filename)
{
StringArray tokens;
StringArray rec;
rec.Read(filename);
// Load estimated per marker error rates
if (rec.Length() == markers)
{
printf(" Updating error rates using data in %s ...\n", (const char *) filename);
for (int i = 0; i < markers; i++)
{
tokens.ReplaceTokens(rec[i+1]);
if (tokens.Length() >= 2) R[i] = tokens[1].AsDouble();
}
return true;
}
return false;
}
void BedFile::openForRead(const char* bedFile, const char* bimFile, const char* famFile, const char* refFile, int nbuf) {
StringArray tokens;
reset();
iFile = ifopen(bedFile,"rb");
if ( iFile == NULL ) {
throw VcfFileException("Failed opening file %s - %s",bedFile,strerror(errno));
}
// read magic numbers
char magicNumbers[3] = {0x6c,0x1b,0x01};
char firstThreeBytes[3];
ifread( iFile, firstThreeBytes, 3 );
for(int i=0; i < 3; ++i) {
if ( firstThreeBytes[i] != magicNumbers[i] ) {
throw VcfFileException("The magic numbers do not match in BED file %s",bedFile);
}
}
iBimFile = ifopen(bimFile,"rb");
iFamFile = ifopen(famFile,"rb");
sRefFile = refFile;
while( 1 ) {
int ret = line.ReadLine(iFamFile);
if ( ret <= 0 ) break;
tokens.ReplaceTokens(line, " \t\r\n");
if ( tokens.Length() < 5 ) {
throw VcfFileException("Less then 5 columns are observed in FAM file");
}
VcfInd* p = new VcfInd(tokens[1],tokens[0],tokens[2],tokens[3],tokens[4]);
vpVcfInds.push_back(p);
}
//Logger::gLogger->writeLog("Finished loading %d individuals from FAM file",(int)vpVcfInds.size());
nBytes = (vpVcfInds.size()+3)/4;
if ( pBedBuffer != NULL ) { delete[] pBedBuffer; }
pBedBuffer = new char[nBytes];
nBuffers = nbuf;
nNumMarkers = 0;
nHead = 0;
bParseGenotypes = true;
bParseDosages = false;
bParseValues = false;
if ( nBuffers == 0 ) { // infinite buffer size
// do not set size of markers
}
else {
vpVcfMarkers.resize( nBuffers );
for(int i=0; i < nBuffers; ++i) {
VcfMarker* p = new VcfMarker;
vpVcfMarkers[i] = p;
}
}
genomeSequence.setReferenceName(sRefFile.c_str());
genomeSequence.useMemoryMap(true);
//Logger::gLogger->writeLog("Loading reference file %s",sRefFile.c_str());
if ( genomeSequence.open() ) {
// write a message that new index file is being created
if ( genomeSequence.create(false) ) {
throw VcfFileException("Failed creating index file of the reference. Please check the file permission");
}
if ( genomeSequence.open() ) {
throw VcfFileException("Failed opening index file of the reference.");
}
}
}
