pos_t VariantProcessor::processAlignment(const BamAlignment& alignment) {
/*
For each alignment, extract the MD and NM tags, validate against
CIGAR string, and create Variants and ReadHaplotypes. All reads
for a block are stored in a deque, and processed again to create
candidate haplotypes.
Returns the start position of this alignment (TODO correct?)
*/
if (!alignment.HasTag("NM") || !alignment.HasTag("MD")) {
std::cerr << "error: BamAlignment '" << alignment.Name <<
"' does not have either NM or MD tags" << std::endl;
}
int nm_tag;
string md_tag;
unsigned int aln_len = alignment.GetEndPosition() - alignment.Position;
alignment.GetTag("MD", md_tag);
alignment.GetTag("NM", nm_tag);
// Reconstruct reference sequence using MD tags
string refseq = createReferenceSequence(alignment);
// With reconstructed reference sequence and query sequence, look
// for variants. It's a bit roundabout to reconstruct reference from
// MD, then use it to find variants (already in MD) but keeping
// state between CIGAR and MD is tricky. This also is a good
// validation; variants found must much the number of variants in
// CIGAR/MD.
vector<VariantPtr> variants;
vector<VariantPtr> read_variants;
const vector<CigarOp>& cigar = alignment.CigarData;
int refpos = 0, readpos = 0;
for (vector<CigarOp>::const_iterator op = cigar.begin(); op != cigar.end(); ++op) {
if (op->Type == 'S') {
readpos += op->Length;
} else if (op->Type == 'M') {
// match or SNP
processMatchOrMismatch(alignment, read_variants, op->Length, refseq, refpos, readpos);
readpos += op->Length;
refpos += op->Length;
} else if (op->Type == 'I') {
processInsertion(alignment, read_variants, op->Length, refseq, refpos, readpos);
readpos += op->Length;
} else if (op->Type == 'D') {
processDeletion(alignment, read_variants, op->Length, refseq, refpos, readpos);
refpos += op->Length; // deletion w.r.t reference; skip ref length
} else {
cerr << "error: unidentified CIGAR type: " << op->Type << endl;
exit(1);
}
}
// Add to alignments list
block_alignments.push_back(alignment);
return 0; // TODO(vsbuffalo)
}
int DataStatisticsTool::Execute()
{
// iterate over reads in BAM file(s)
BamAlignment alignObj;
while(bamReader.GetNextAlignment(alignObj))
{
if (alignObj.IsDuplicate()) continue;
if (alignObj.IsFailedQC()) continue;
if (!alignObj.IsMapped()) continue;
if (!alignObj.IsPrimaryAlignment()) continue;
if (alignObj.IsPaired() && !alignObj.IsProperPair()) continue;
if (alignObj.IsPaired() && !alignObj.IsMateMapped()) continue;
if (!alignObj.HasTag("MD")) continue;
// // debug
// GenericBamAlignmentTools::printBamAlignmentCigar(alignObj);
// GenericBamAlignmentTools::printBamAlignmentMD(alignObj);
// shift InDel
GenericBamAlignmentTools::leftShiftInDel(alignObj);
// // debug
// GenericBamAlignmentTools::printBamAlignmentCigar(alignObj);
// GenericBamAlignmentTools::printBamAlignmentMD(alignObj);
// get the alignment sequences
string alignRead;
string alignGenome;
GenericBamAlignmentTools::getAlignmentSequences(alignObj, alignRead, alignGenome);
// update the statistics
statistics.update(alignRead, alignGenome);
}
// print to screen
cout << statistics << endl;
// statistics.printMatchMismatch();
// close BAM reader
bamReader.Close();
// close Fasta
genomeFasta.Close();
return 1;
}
int main (int argc, char *argv[]) {
if( (argc== 1) ||
(argc== 2 && string(argv[1]) == "-h") ||
(argc== 2 && string(argv[1]) == "-help") ||
(argc== 2 && string(argv[1]) == "--help") ){
cout<<"Usage:editDist [in bam]"<<endl<<"this program returns the NM field of all aligned reads"<<endl;
return 1;
}
string bamfiletopen = string(argv[1]);
// cout<<bamfiletopen<<endl;
BamReader reader;
// cout<<"ok"<<endl;
if ( !reader.Open(bamfiletopen) ) {
cerr << "Could not open input BAM files." << endl;
return 1;
}
BamAlignment al;
// cout<<"ok"<<endl;
while ( reader.GetNextAlignment(al) ) {
// cout<<al.Name<<endl;
if(!al.IsMapped())
continue;
if(al.HasTag("NM") ){
int editDist;
if(al.GetTag("NM",editDist) ){
cout<<editDist<<endl;
}else{
cerr<<"Cannot retrieve NM field for "<<al.Name<<endl;
return 1;
}
}else{
cerr<<"Warning: read "<<al.Name<<" is aligned but has no NM field"<<endl;
}
} //while al
reader.Close();
return 0;
}
//{{{ void process_intra_chrom_split(const BamAlignment &curr,
void
SV_SplitRead::
process_intra_chrom_split(const BamAlignment &curr,
const RefVector refs,
BamWriter &inter_chrom_reads,
map<string, BamAlignment> &mapped_splits,
UCSCBins<SV_BreakPoint*> &r_bin,
int weight,
int id,
int sample_id,
SV_SplitReadReader *_reader)
{
if (mapped_splits.find(curr.Name) == mapped_splits.end()) {
uint32_t clipped = count_clipped(curr.CigarData);
if ( curr.HasTag("YP") == true) {
uint32_t t;
curr.GetTag("YP", t);
if (t == 2)
mapped_splits[curr.Name] = curr;
}
else if (clipped >= _reader->min_clip)
mapped_splits[curr.Name] = curr;
} else {
if ( mapped_splits[curr.Name].RefID == curr.RefID ) {
try {
SV_SplitRead *new_split_read =
new SV_SplitRead(mapped_splits[curr.Name],
curr,
refs,
weight,
id,
sample_id,
_reader);
SV_BreakPoint *new_bp = NULL;
if (new_split_read->is_sane()) {
new_bp = new_split_read->get_bp();
if (new_bp != NULL) {
new_bp->cluster(r_bin);
} else {
cerr << "Alignment name:" << curr.Name << endl;
free(new_split_read);
}
} else
free(new_split_read);
} catch (int) {
cerr << "Error creating split read: " << endl;
}
} else {
BamAlignment al1 = curr;
BamAlignment al2 = mapped_splits[curr.Name];
al1.MateRefID = al2.RefID;
al2.MateRefID = al1.RefID;
al1.MatePosition = al2.Position;
al2.MatePosition = al1.Position;
string x = _reader->get_source_file_name();
al1.AddTag("LS","Z",x);
al2.AddTag("LS","Z",x);
inter_chrom_reads.SaveAlignment(al1);
inter_chrom_reads.SaveAlignment(al2);
}
mapped_splits.erase(curr.Name);
}
}
int main (int argc, char *argv[]) {
if( (argc!= 3) ||
(argc== 2 && string(argv[1]) == "-h") ||
(argc== 2 && string(argv[1]) == "-help") ||
(argc== 2 && string(argv[1]) == "--help") ){
cerr<<"Usage:splitByRG [in bam] [out prefix]"<<endl<<"this program creates one bam file per RG in the with the outprefix\nFor example splitByRG in.bam out will create\nout.rg1.bam\nout.rg2.bam\n"<<endl;
return 1;
}
string bamfiletopen = string(argv[1]);
// if(!strEndsWith(bamfiletopen,".bam")){
// }
string bamDirOutPrefix = string(argv[2]);
map<string,BamWriter *> rg2BamWriter;
// if(!isDirectory(bamDirOut)){
// cerr<<"ERROR: the out directory does not exist"<<endl;
// return 1;
// }
BamReader reader;
if ( !reader.Open(bamfiletopen) ) {
cerr << "Could not open input BAM files." << endl;
return 1;
}
SamHeader header = reader.GetHeader();
const RefVector references = reader.GetReferenceData();
vector<RefData> refData=reader.GetReferenceData();
string pID = "splitByRG";
string pName = "splitByRG";
string pCommandLine = "";
for(int i=0;i<(argc);i++){
pCommandLine += (string(argv[i])+" ");
}
putProgramInHeader(&header,pID,pName,pCommandLine,returnGitHubVersion(string(argv[0]),".."));
SamReadGroupDictionary srgd=header.ReadGroups;
for(SamReadGroupConstIterator srgci=srgd.ConstBegin();
srgci<srgd.ConstEnd();
srgci++){
//cout<<*srgci<<endl;
const SamReadGroup rg = (*srgci);
//cout<<rg.ID<<endl;
rg2BamWriter[rg.ID] = new BamWriter();
rg2BamWriter[rg.ID]->Open(bamDirOutPrefix+"."+rg.ID+".bam",header,references);
}
BamAlignment al;
unsigned int total=0;
while ( reader.GetNextAlignment(al) ) {
// al.SetIsFailedQC(false);
// writer.SaveAlignment(al);
// if(al.IsMapped () ){
// if(rg2BamWriter.find(refData[al.RefID].RefName) == rg2BamWriter.end()){ //new
// rg2BamWriter[refData[al.RefID].RefName] = new BamWriter();
// if ( !rg2BamWriter[refData[al.RefID].RefName]->Open(bamDirOutPrefix+"."+refData[al.RefID].RefName+".bam",header,references) ) {
// cerr << "Could not open output BAM file "<< bamDirOutPrefix<<"."<<refData[al.RefID].RefName<<".bam" << endl;
// return 1;
// }
// }else{
// rg2BamWriter[refData[al.RefID].RefName]->SaveAlignment(al);
// }
// }else{
// unmapped.SaveAlignment(al);
// }
if(al.HasTag("RG")){
string rgTag;
al.GetTag("RG",rgTag);
//cout<<rgTag<<endl;
if(rg2BamWriter.find(rgTag) == rg2BamWriter.end()){ //new
cerr<<"Found new RG "<<rgTag<<endl;
rg2BamWriter[rgTag] = new BamWriter();
if ( !rg2BamWriter[rgTag]->Open(bamDirOutPrefix+"."+rgTag+".bam",header,references) ) {
cerr << "Could not open output BAM file "<< bamDirOutPrefix<<"."<<rgTag<<".bam" << endl;
return 1;
}
rg2BamWriter[rgTag]->SaveAlignment(al);
}else{
rg2BamWriter[rgTag]->SaveAlignment(al);
}
}else{
string rgTag="unknown";
//cout<<rgTag<<endl;
if(rg2BamWriter.find(rgTag) == rg2BamWriter.end()){ //new
cerr<<"Found new RG "<<rgTag<<endl;
rg2BamWriter[rgTag] = new BamWriter();
if ( !rg2BamWriter[rgTag]->Open(bamDirOutPrefix+"."+rgTag+".bam",header,references) ) {
cerr << "Could not open output BAM file "<< bamDirOutPrefix<<"."<<rgTag<<".bam" << endl;
return 1;
}
rg2BamWriter[rgTag]->SaveAlignment(al);
}else{
rg2BamWriter[rgTag]->SaveAlignment(al);
}
// cerr << "Cannot get RG tag for " << al.Name<<endl;
// return 1;
}
total++;
} //while al
reader.Close();
// writer.Close();
// unmapped.Close();
map<string,BamWriter *>::iterator rg2BamWriterIt;
for (rg2BamWriterIt =rg2BamWriter.begin();
rg2BamWriterIt!=rg2BamWriter.end();
rg2BamWriterIt++){
rg2BamWriterIt->second->Close();
}
cerr<<"Wrote succesfully "<<total<<" reads"<<endl;
return 0;
}
int main (int argc, char *argv[]) {
if( (argc== 1) ||
(argc== 2 && string(argv[1]) == "-h") ||
(argc== 2 && string(argv[1]) == "-help") ||
(argc== 2 && string(argv[1]) == "--help") ){
cout<<"Usage:"<<endl;
cout<<""<<endl;
cout<<"plotQualScore input.bam"<<endl;
return 1;
}
string bamfiletopen = string(argv[1]);
BamReader reader;
if ( !reader.Open(bamfiletopen) ) {
cerr << "Could not open input BAM files." << endl;
return 1;
}
// if ( !reader.LocateIndex() ){
// cerr << "warning: cannot locate index for file " << bamfiletopen<<endl;
// //return 1;
// }
BamAlignment al;
BamAlignment al2;
bool unsurePEorSE=true;
bool pe=true;
int strLength=-1;
int vecLengthToUse=-1;
map<short,unsigned long> ** counterA = 0;
map<short,unsigned long> ** counterC = 0;
map<short,unsigned long> ** counterG = 0;
map<short,unsigned long> ** counterT = 0;
int lengthIndex1=0;
int lengthIndex2=0;
string seqInd1;
string seqInd2;
string qualInd1;
string qualInd2;
int offsetInd2;
while ( reader.GetNextAlignment(al) ) {
if(unsurePEorSE){
strLength=al.QueryBases.length();
if(al.IsPaired()){
pe=true;
vecLengthToUse=2*strLength;
}else{
pe=false;
vecLengthToUse=strLength;
}
string index1;
string index2;
if(al.HasTag("XI")){
al.GetTag("XI",index1);
vecLengthToUse+=index1.length();
lengthIndex1=index1.length();
}
if(al.HasTag("XJ")){
al.GetTag("XJ",index2);
vecLengthToUse+=index2.length();
lengthIndex2=index2.length();
}
counterA = new map<short,unsigned long> * [vecLengthToUse];
counterC = new map<short,unsigned long> * [vecLengthToUse];
counterG = new map<short,unsigned long> * [vecLengthToUse];
counterT = new map<short,unsigned long> * [vecLengthToUse];
for(int i=0;i<vecLengthToUse;i++){
counterA[i]=new map<short,unsigned long> ();
counterC[i]=new map<short,unsigned long> ();
counterG[i]=new map<short,unsigned long> ();
counterT[i]=new map<short,unsigned long> ();
for(short k=minQualScore;k<=maxQualScore;k++){
(*counterA[i])[k]=0;
(*counterC[i])[k]=0;
(*counterG[i])[k]=0;
(*counterT[i])[k]=0;
}
}
unsurePEorSE=false;
}else{
if(pe &&
!al.IsPaired()){
cerr << "Cannot have unpaired reads in PE mode" << endl;
return 1;
}
if(!pe &&
al.IsPaired()){
cerr << "Cannot have unpaired reads in SE mode" << endl;
return 1;
}
}
if(al.QueryBases.length() != al.Qualities.length()){
cerr << "Cannot have different lengths for sequence and quality" << endl;
return 1;
}
if(int(al.QueryBases.length()) != strLength){
cerr << "Cannot have different lengths for sequence and quality" << endl;
return 1;
}
if(pe){
if(al.IsFirstMate()){
reader.GetNextAlignment(al2);
if(al2.QueryBases.length() != al2.Qualities.length()){
cerr << "Cannot have different lengths for sequence and quality" << endl;
return 1;
}
}else{
cerr << "First read should be the first mate" << endl;
return 1;
}
}
//cycle
for(unsigned int i=0;i<al.QueryBases.length();i++){
short x=(short(al.Qualities[i])-qualOffset);
if(al.QueryBases[i] == 'A'){
(*counterA[i])[x]++;
}
if(al.QueryBases[i] == 'C'){
(*counterC[i])[x]++;
}
if(al.QueryBases[i] == 'G'){
(*counterG[i])[x]++;
}
if(al.QueryBases[i] == 'T'){
(*counterT[i])[x]++;
}
}
//The indices for al and al2 should hopefully be the same
if(lengthIndex1>0){
al.GetTag("XI",seqInd1);
al.GetTag("YI",qualInd1);
int j;
for(int i=0;i<lengthIndex1;i++){
j=i+al.QueryBases.length();
short x=(short(qualInd1[i])-qualOffset);
if(seqInd1[i] == 'A'){
(*counterA[j])[x]++;
}
if(seqInd1[i] == 'C'){
(*counterC[j])[x]++;
}
if(seqInd1[i] == 'G'){
(*counterG[j])[x]++;
}
if(seqInd1[i] == 'T'){
(*counterT[j])[x]++;
}
}
}
if(pe){
offsetInd2=al.QueryBases.length()+lengthIndex1+al2.QueryBases.length();
int j;
for(unsigned int i=0;i<al2.QueryBases.length();i++){
j=i+al.QueryBases.length()+lengthIndex1;
short x=(short(al2.Qualities[i])-qualOffset);
if(al2.QueryBases[i] == 'A'){
(*counterA[j])[x]++;
}
if(al2.QueryBases[i] == 'C'){
(*counterC[j])[x]++;
}
if(al2.QueryBases[i] == 'G'){
(*counterG[j])[x]++;
}
if(al2.QueryBases[i] == 'T'){
(*counterT[j])[x]++;
}
}
}else{
offsetInd2=al.QueryBases.length()+lengthIndex1;
}
//The indices for al and al2 should hopefully be the same
if(lengthIndex2>0){
al.GetTag("XJ",seqInd2);
al.GetTag("YJ",qualInd2);
int j;
for(int i=0;i<lengthIndex2;i++){
j=offsetInd2+i;
short x=(short(qualInd2[i])-qualOffset);
if(seqInd2[i] == 'A'){
(*counterA[j])[x]++;
}
if(seqInd2[i] == 'C'){
(*counterC[j])[x]++;
}
if(seqInd2[i] == 'G'){
(*counterG[j])[x]++;
}
if(seqInd2[i] == 'T'){
(*counterT[j])[x]++;
}
}
}
}
reader.Close();
cout<<"cycle\t"<<"nuc\t";
for(short k=minQualScore;k<maxQualScore;k++){
cout<<k<<"\t";
}
cout<<maxQualScore<<endl;
for(int i=0;i<vecLengthToUse;i++){
cout<<(i+1)<<"\t";
cout<<"A\t";
for(short k=minQualScore;k<maxQualScore;k++){
cout<<(*counterA[i])[k]<<"\t";
}
cout<<(*counterA[i])[maxQualScore]<<endl;
cout<<(i+1)<<"\t";
cout<<"C\t";
for(short k=minQualScore;k<maxQualScore;k++){
cout<<(*counterC[i])[k]<<"\t";
}
cout<<(*counterC[i])[maxQualScore]<<endl;
cout<<(i+1)<<"\t";
cout<<"G\t";
for(short k=minQualScore;k<maxQualScore;k++){
cout<<(*counterG[i])[k]<<"\t";
}
cout<<(*counterG[i])[maxQualScore]<<endl;
cout<<(i+1)<<"\t";
cout<<"T\t";
for(short k=minQualScore;k<maxQualScore;k++){
cout<<(*counterT[i])[k]<<"\t";
}
cout<<(*counterT[i])[maxQualScore]<<endl;
}
return 0;
}
int main (int argc, char *argv[]) {
bool produceUnCompressedBAM=false;
bool verbose=false;
bool ancientDNA=false;
bool keepOrig=false;
string adapter_F=options_adapter_F_BAM;
string adapter_S=options_adapter_S_BAM;
string adapter_chimera=options_adapter_chimera_BAM;
string key="";
bool allowMissing=false;
int trimCutoff=1;
bool allowAligned=false;
bool printLog=false;
string logFileName;
BamReader reader;
BamWriter writer;
string bamFile;
string bamFileOUT="";
string key1;
string key2;
bool useDist=false;
double location=-1.0;
double scale =-1.0;
bool fastqFormat=false;
string fastqfile1 = "";
string fastqfile2 = "";
string fastqoutfile = "";
bool singleEndModeFQ=true;
const string usage=string(string(argv[0])+
" [options] BAMfile"+"\n"+
"\nThis program takes an unaligned BAM where mates are consecutive\nor fastq files and trims and merges reads\n"+
"\n\tYou can specify a unaligned bam file or one or two fastq :\n"+
"\t\t"+"-fq1" +"\t\t"+"First fastq"+"\n"+
"\t\t"+"-fq2" +"\t\t"+"Second fastq file (for paired-end)"+"\n"+
"\t\t"+"-fqo" +"\t\t"+"Output fastq prefix"+"\n\n"+
//"\t"+"-p , --PIPE"+"\n\t\t"+"Read BAM from and write it to PIPE"+"\n"+
"\t"+"-o , --outfile" +"\t\t"+"Output (BAM format)."+"\n"+
"\t"+"-u " +"\t\t"+"Produce uncompressed bam (good for pipe)"+"\n"+
// "\t"+" , --outprefix" +"\n\t\t"+"Prefix for output files (default '"+outprefix+"')."+"\n"+
//"\t"+" , --SAM" +"\n\t\t"+"Output SAM not BAM."+"\n"+
"\t"+"--aligned" +"\t\t"+"Allow reads to be aligned (default "+boolStringify(allowAligned)+")"+"\n"+
"\t"+"-v , --verbose" +"\t\t"+"Turn all messages on (default "+boolStringify(verbose)+")"+"\n"+
"\t"+"--log [log file]" +"\t"+"Print a tally of merged reads to this log file (default only to stderr)"+"\n"+
"\n\t"+"Paired End merging/Single Read trimming options"+"\n"+
"\t\t"+"You can specify either:"+"\n"+
"\t\t\t"+"--ancientdna"+"\t\t\t"+"ancient DNA (default "+boolStringify(ancientDNA)+")"+"\n"+
"\t\t"+" "+"\t\t\t\t"+"this allows for partial overlap"+"\n"+
"\n\t\t"+"or if you know your size length distribution:"+"\n"+
"\t\t\t"+"--loc"+"\t\t\t\t"+"Location for lognormal dist. (default none)"+"\n"+
"\t\t\t"+"--scale"+"\t\t\t\t"+"Scale for lognormal dist. (default none)"+"\n"+
// "\t\t\t\t\t\t\tGood for merging ancient DNA reads into a single sequence\n\n"
"\n\t\t"+"--keepOrig"+"\t\t\t\t"+"Write original reads if they are trimmed or merged (default "+boolStringify(keepOrig)+")"+"\n"+
"\t\t\t\t\t\t\tSuch reads will be marked as PCR duplicates\n\n"
"\t\t"+"-f , --adapterFirstRead" +"\t\t\t"+"Adapter that is observed after the forward read (def. Multiplex: "+options_adapter_F_BAM.substr(0,30)+")"+"\n"+
"\t\t"+"-s , --adapterSecondRead" +"\t\t"+"Adapter that is observed after the reverse read (def. Multiplex: "+options_adapter_S_BAM.substr(0,30)+")"+"\n"+
"\t\t"+"-c , --FirstReadChimeraFilter" +"\t\t"+"If the forward read looks like this sequence, the cluster is filtered out.\n\t\t\t\t\t\t\tProvide several sequences separated by comma (def. Multiplex: "+options_adapter_chimera_BAM.substr(0,30)+")"+"\n"+
"\t\t"+"-k , --key"+"\t\t\t\t"+"Key sequence with which each sequence starts. Comma separate for forward and reverse reads. (default '"+key+"')"+"\n"+
"\t\t"+"-i , --allowMissing"+"\t\t\t"+"Allow one base in one key to be missing or wrong. (default "+boolStringify(allowMissing)+")"+"\n"+
"\t\t"+"-t , --trimCutoff"+"\t\t\t"+"Lowest number of adapter bases to be observed for single Read trimming (default "+stringify(trimCutoff)+")");
if( (argc== 1) ||
(argc== 2 && string(argv[1]) == "-h") ||
(argc== 2 && string(argv[1]) == "-help") ||
(argc== 2 && string(argv[1]) == "--help") ){
cout<<"Usage:"<<endl;
cout<<""<<endl;
cout<<usage<<endl;
return 1;
}
for(int i=1;i<(argc-1);i++){ //all but the last arg
if(strcmp(argv[i],"-fq1") == 0 ){
fastqfile1=string(argv[i+1]);
fastqFormat=true;
i++;
continue;
}
if(strcmp(argv[i],"-fq2") == 0 ){
fastqfile2=string(argv[i+1]);
fastqFormat=true;
singleEndModeFQ=false;
i++;
continue;
}
if(strcmp(argv[i],"-fqo") == 0 ){
fastqoutfile=string(argv[i+1]);
fastqFormat=true;
i++;
continue;
}
if(strcmp(argv[i],"--log") == 0 ){
logFileName =string(argv[i+1]);
printLog=true;
i++;
continue;
}
if(strcmp(argv[i],"-p") == 0 || strcmp(argv[i],"--PIPE") == 0 ){
cerr<<"This version no longer works with pipe, exiting"<<endl;
return 1;
}
if(strcmp(argv[i],"-u") == 0 ){
produceUnCompressedBAM=true;
continue;
}
if(strcmp(argv[i],"--aligned") == 0 ){
allowAligned=true;
continue;
}
if(strcmp(argv[i],"-o") == 0 || strcmp(argv[i],"--outfile") == 0 ){
bamFileOUT =string(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"-v") == 0 || strcmp(argv[i],"--verbose") == 0 ){
verbose=true;
continue;
}
if(strcmp(argv[i],"--ancientdna") == 0 ){
ancientDNA=true;
continue;
}
if(strcmp(argv[i],"--keepOrig") == 0 ){
keepOrig=true;
continue;
}
if(strcmp(argv[i],"--loc") == 0 ){
location =destringify<double>(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"--scale") == 0 ){
scale =destringify<double>(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"-f") == 0 || strcmp(argv[i],"--adapterFirstRead") == 0 ){
adapter_F =string(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"-s") == 0 || strcmp(argv[i],"--adapterSecondRead") == 0 ){
adapter_S =string(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"-c") == 0 || strcmp(argv[i],"--FirstReadChimeraFilter") == 0 ){
adapter_chimera =string(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"-k") == 0 || strcmp(argv[i],"--keys") == 0 ){
key =string(argv[i+1]);
i++;
continue;
}
if(strcmp(argv[i],"-i") == 0 || strcmp(argv[i],"--allowMissing") == 0 ){
allowMissing=true;
continue;
}
if(strcmp(argv[i],"-t") == 0 || strcmp(argv[i],"--trimCutoff") == 0 ){
trimCutoff=atoi(argv[i+1]);
i++;
continue;
}
cerr<<"Unknown option "<<argv[i] <<" exiting"<<endl;
return 1;
}
bamFile=argv[argc-1];
if( (location != -1.0 && scale == -1.0) ||
(location == -1.0 && scale != -1.0) ){
cerr<<"Cannot specify --location without specifying --scale"<<endl;
return 1;
}
if( (location != -1.0 && scale != -1.0) ){
useDist=true;
if(ancientDNA){
cerr<<"Cannot specify --location/--scale and --ancientDNA"<<endl;
return 1;
}
}
MergeTrimReads mtr (adapter_F,adapter_S,adapter_chimera,
key1,key2,
trimCutoff,allowMissing,ancientDNA,location,scale,useDist);
fqwriters onereadgroup;
if(fastqFormat){
if( bamFileOUT != "" || produceUnCompressedBAM || allowAligned){
cerr<<"ERROR : Cannot specify options like -o, -u or --allowAligned for fastq"<<endl;
return 1;
}
if(fastqfile1 == ""){
cerr<<"ERROR : Must specify as least the first file for fastq"<<endl;
return 1;
}
FastQParser * fqp1;
FastQParser * fqp2;
if(singleEndModeFQ){
fqp1 = new FastQParser (fastqfile1);
string outdirs = fastqoutfile+".fq.gz";
string outdirsf = fastqoutfile+".fail.fq.gz";
onereadgroup.single.open(outdirs.c_str(), ios::out);
onereadgroup.singlef.open(outdirsf.c_str(), ios::out);
if(!onereadgroup.single.good()){ cerr<<"Cannot write to file "<<outdirs<<endl; return 1; }
if(!onereadgroup.singlef.good()){ cerr<<"Cannot write to file "<<outdirsf<<endl; return 1; }
}else{
fqp1 = new FastQParser (fastqfile1);
fqp2 = new FastQParser (fastqfile2);
string outdirs = fastqoutfile+".fq.gz";
string outdir1 = fastqoutfile+"_r1.fq.gz";
string outdir2 = fastqoutfile+"_r2.fq.gz";
string outdirsf = fastqoutfile+".fail.fq.gz";
string outdir1f = fastqoutfile+"_r1.fail.fq.gz";
string outdir2f = fastqoutfile+"_r2.fail.fq.gz";
onereadgroup.single.open(outdirs.c_str(), ios::out);
onereadgroup.pairr1.open(outdir1.c_str(), ios::out);
onereadgroup.pairr2.open(outdir2.c_str(), ios::out);
onereadgroup.singlef.open(outdirsf.c_str(), ios::out);
onereadgroup.pairr1f.open(outdir1f.c_str(), ios::out);
onereadgroup.pairr2f.open(outdir2f.c_str(), ios::out);
if(!onereadgroup.single.good()){ cerr<<"Cannot write to file "<<outdirs<<endl; return 1; }
if(!onereadgroup.pairr1.good()){ cerr<<"Cannot write to file "<<outdir1<<endl; return 1; }
if(!onereadgroup.pairr2.good()){ cerr<<"Cannot write to file "<<outdir2<<endl; return 1; }
if(!onereadgroup.singlef.good()){ cerr<<"Cannot write to file "<<outdirsf<<endl; return 1; }
if(!onereadgroup.pairr1f.good()){ cerr<<"Cannot write to file "<<outdir1f<<endl; return 1; }
if(!onereadgroup.pairr2f.good()){ cerr<<"Cannot write to file "<<outdir2f<<endl; return 1; }
}
unsigned int totalSeqs=0;
while(fqp1->hasData()){
FastQObj * fo1=fqp1->getData();
vector<string> def1=allTokens( *(fo1->getID()), ' ' );
string def1s=def1[0];
FastQObj * fo2;
string def2s;
string ext2s;
if(!singleEndModeFQ){
if(!fqp2->hasData()){
cerr << "ERROR: Discrepency between fastq files at record " << *(fo1->getID()) <<endl;
return 1;
}
fo2=fqp2->getData();
vector<string> def2=allTokens( *(fo2->getID()), ' ' );
def2s=def2[0];
if(strEndsWith(def1s,"/1")){
def1s=def1s.substr(0,def1s.size()-2);
}
if(strEndsWith(def2s,"/2")){
def2s=def2s.substr(0,def2s.size()-2);
}
if(strBeginsWith(def1s,"@")){
def1s=def1s.substr(1,def1s.size()-1);
}
if(strBeginsWith(def2s,"@")){
def2s=def2s.substr(1,def2s.size()-1);
}
if(def1s != def2s){
cerr << "ERROR: Discrepency between fastq files, different names " << *(fo1->getID()) <<" and "<< *(fo2->getID()) <<endl;
return 1;
}
merged result= mtr.process_PE(*(fo1->getSeq()),*(fo1->getQual()),
*(fo2->getSeq()),*(fo2->getQual()));
mtr.incrementCountall();
if(result.code != ' '){ //keys or chimeras
if(result.code == 'K'){
mtr.incrementCountfkey();
}else{
if(result.code == 'D'){
mtr.incrementCountchimera();
}else{
cerr << "leehom: Wrong return code =\""<<result.code<<"\""<<endl;
exit(1);
}
}
onereadgroup.pairr2f<<"@"<<def2s<<"/2" <<endl <<*(fo2->getSeq())<<endl<<"+"<<endl <<*(fo2->getQual())<<endl;
onereadgroup.pairr1f<<"@"<<def1s<<"/1" <<endl <<*(fo1->getSeq())<<endl<<"+"<<endl <<*(fo1->getQual())<<endl;
continue;
}else{
if(result.sequence != ""){ //new sequence
onereadgroup.single<<"@"<<def1s<<"" <<endl << result.sequence<<endl<<"+"<<endl <<result.quality<<endl;
if( result.sequence.length() > max(fo1->getSeq()->length(),fo2->getSeq()->length()) ){
mtr.incrementCountmergedoverlap();
}else{
mtr.incrementCountmerged();
}
}else{ //keep as is
mtr.incrementCountnothing();
onereadgroup.pairr2<<"@"<<def2s<<"/2" <<endl <<*(fo2->getSeq())<<endl<<"+"<<endl <<*(fo2->getQual())<<endl;
onereadgroup.pairr1<<"@"<<def1s<<"/1" <<endl <<*(fo1->getSeq())<<endl<<"+"<<endl <<*(fo1->getQual())<<endl;
}
}
}else{
merged result=mtr.process_SR(*(fo1->getSeq()),*(fo1->getQual()));
mtr.incrementCountall();
if(result.code != ' '){ //either chimera or missing key
if(result.code == 'K'){
mtr.incrementCountfkey();
}else{
if(result.code == 'D'){
mtr.incrementCountchimera();
}else{
cerr << "leehom: Wrong return code =\""<<result.code<<"\""<<endl;
exit(1);
}
}
onereadgroup.singlef<<""<<*(fo1->getID())<<"" <<endl << *(fo1->getSeq())<<endl<<"+"<<endl <<*(fo1->getQual())<<endl;
continue;
}
if(result.sequence != ""){ //new sequence
mtr.incrementCounttrimmed();
onereadgroup.single<<""<<*(fo1->getID())<<"" <<endl << result.sequence<<endl<<"+"<<endl <<result.quality<<endl;
}else{
mtr.incrementCountnothing();
onereadgroup.single<<""<<*(fo1->getID())<<"" <<endl << *(fo1->getSeq())<<endl<<"+"<<endl <<*(fo1->getQual())<<endl;
}
}
totalSeqs++;
}
delete fqp1;
if(!singleEndModeFQ){
delete fqp2;
}
if(singleEndModeFQ){
onereadgroup.single.close();
onereadgroup.singlef.close();
}else{
onereadgroup.single.close();
onereadgroup.pairr1.close();
onereadgroup.pairr2.close();
onereadgroup.singlef.close();
onereadgroup.pairr1f.close();
onereadgroup.pairr2f.close();
}
//fastq
}else{
//else BAM
// initMerge();
// set_adapter_sequences(adapter_F,
// adapter_S,
// adapter_chimera);
// set_options(trimCutoff,allowMissing,mergeoverlap);
if(key != ""){
size_t found=key.find(",");
if (found == string::npos){ //single end reads
key1=key;
key2="";
} else{ //paired-end
key1=key.substr(0,found);
key2=key.substr(found+1,key.length()-found+1);
}
}
if( bamFileOUT == "" ){
cerr<<"The output must be a be specified, exiting"<<endl;
return 1;
}
if ( !reader.Open(bamFile) ) {
cerr << "Could not open input BAM file "<<bamFile << endl;
return 1;
}
SamHeader header = reader.GetHeader();
string pID = "mergeTrimReadsBAM";
string pName = "mergeTrimReadsBAM";
string pCommandLine = "";
for(int i=0;i<(argc);i++){
pCommandLine += (string(argv[i])+" ");
}
putProgramInHeader(&header,pID,pName,pCommandLine,returnGitHubVersion(string(argv[0]),".."));
const RefVector references = reader.GetReferenceData();
//we will not call bgzip with full compression, good for piping into another program to
//lessen the load on the CPU
if(produceUnCompressedBAM)
writer.SetCompressionMode(BamWriter::Uncompressed);
if ( !writer.Open(bamFileOUT,header,references) ) {
cerr << "Could not open output BAM file "<<bamFileOUT << endl;
return 1;
}
SamHeader sh=reader.GetHeader();
//Up to the user to be sure that a sequence is followed by his mate
// if(!sh.HasSortOrder() ||
// sh.SortOrder != "queryname"){
// cerr << "Bamfile must be sorted by queryname" << endl;
// return 1;
// }
BamAlignment al;
BamAlignment al2;
bool al2Null=true;
while ( reader.GetNextAlignment(al) ) {
if(al.IsMapped() || al.HasTag("NM") || al.HasTag("MD") ){
if(!allowAligned){
cerr << "Reads should not be aligned" << endl;
return 1;
}else{
//should we remove tags ?
}
}
if(al.IsPaired() &&
al2Null ){
al2=al;
al2Null=false;
continue;
}else{
if(al.IsPaired() &&
!al2Null){
bool result = mtr.processPair(al,al2);
if( result ){//was merged
BamAlignment orig;
BamAlignment orig2;
if(keepOrig){
orig2 = al2;
orig = al;
}
writer.SaveAlignment(al);
if(keepOrig){
orig.SetIsDuplicate(true);
orig2.SetIsDuplicate(true);
writer.SaveAlignment(orig2);
writer.SaveAlignment(orig);
}
//the second record is empty
}else{
//keep the sequences as pairs
writer.SaveAlignment(al2);
writer.SaveAlignment(al);
}
//
// SINGLE END
//
}else{
BamAlignment orig;
if(keepOrig){
orig =al;
}
mtr.processSingle(al);
if(keepOrig){
//write duplicate
if(orig.QueryBases.length() != al.QueryBases.length()){
orig.SetIsDuplicate(true);
writer.SaveAlignment(orig);
}
}
writer.SaveAlignment(al);
} //end single end
al2Null=true;
}//second pair
} //while al
reader.Close();
writer.Close();
} //else BAM
cerr <<mtr.reportSingleLine()<<endl;
if(printLog){
ofstream fileLog;
fileLog.open(logFileName.c_str());
if (fileLog.is_open()){
fileLog <<mtr.reportMultipleLines() <<endl;
}else{
cerr << "Unable to print to file "<<logFileName<<endl;
}
fileLog.close();
}
return 0;
}
int main (int argc, char *argv[]) {
if( (argc!= 4 && argc !=5 && argc !=6) ||
(argc== 2 && string(argv[1]) == "-h") ||
(argc== 2 && string(argv[1]) == "-help") ||
(argc== 2 && string(argv[1]) == "--help") ){
cerr<<"Usage:splitByRG [in bam] [rg Tally] [out prefix] (optional target)"<<endl<<"this program will subsample a BAM file per read group for a certain target\nFor example splitByRG in.bam tally.txt out will create\nout.rg1.bam\nout.rg2.bam\n"<<endl;
return 1;
}
string bamfiletopen = string(argv[1]);
string rgTally = string(argv[2]);
string bamDirOutPrefix = string(argv[3]);
int target = 200000;
int maxTarget = 1000000;
if(argc==5){
target = destringify<int> ( string(argv[4]) );
}
if(argc==6){
target = destringify<int> ( string(argv[4]) );
maxTarget = destringify<int> ( string(argv[5]) );
}
cerr<<"minimum fragments:\t"<<target<<endl;
cerr<<"target fragments:\t"<<maxTarget<<endl;
string line;
ifstream myFileTally;
map<string,double> rg2Fraction;
myFileTally.open(rgTally.c_str(), ios::in);
cerr<<"Retained groups:\n"<<endl;
cerr<<"RG\t#mapped\tfraction retained"<<endl;
cerr<<"-----------------------------------"<<endl;
if (myFileTally.is_open()){
while ( getline (myFileTally,line)){
vector<string> tokens = allTokens(line,'\t');
if(tokens.size() > 6)
if( tokens[1] == "pass" &&
(tokens[0] != "\"\"" &&
tokens[0] != "control" &&
tokens[0] != "TOTAL") ){
//cout<<tokens[0]<<"\t"<<tokens[5]<<endl;
int count = destringify<int>(tokens[5]);
if(count>target){
if(count>=maxTarget){
rg2Fraction[ tokens[0] ] = double(maxTarget)/double(count);
cout<<tokens[0]<<"\t"<<count<<"\t"<<double(maxTarget)/double(count)<<endl;
}else{
cout<<tokens[0]<<"\t"<<count<<"\t"<<1.0<<endl;
rg2Fraction[ tokens[0] ] = 1.0;
}
}
}
}
myFileTally.close();
}else{
cerr << "Unable to open file "<<rgTally<<endl;
return 1;
}
map<string,BamWriter *> rg2BamWriter;
// if(!isDirectory(bamDirOut)){
// cerr<<"ERROR: the out directory does not exist"<<endl;
// return 1;
// }
BamReader reader;
if ( !reader.Open(bamfiletopen) ) {
cerr << "Could not open input BAM files." << endl;
return 1;
}
const SamHeader header = reader.GetHeader();
const RefVector references = reader.GetReferenceData();
vector<RefData> refData=reader.GetReferenceData();
SamReadGroupDictionary srgd=header.ReadGroups;
for(SamReadGroupConstIterator srgci=srgd.ConstBegin();
srgci<srgd.ConstEnd();
srgci++){
//cout<<*srgci<<endl;
const SamReadGroup rg = (*srgci);
//cout<<rg.ID<<endl;
if( rg2Fraction.find(rg.ID) != rg2Fraction.end() ){
rg2BamWriter[rg.ID] = new BamWriter();
rg2BamWriter[rg.ID]->Open(bamDirOutPrefix+"."+rg.ID+".bam",header,references);
}
//cout<<bamDirOutPrefix+"."+rg.ID+".bam"<<endl;
}
// return 1;
// BamWriter unmapped;
// cout<<header.ToString()<<endl;
// return 1;
// if ( !unmapped.Open(bamDirOutPrefix+".unmapped.bam",header,references) ) {
// cerr << "Could not open output BAM file "<< bamDirOutPrefix+".unmapped.bam" << endl;
// return 1;
// }
// cout<<"reading"<<endl;
BamAlignment al;
unsigned int total=0;
while ( reader.GetNextAlignment(al) ) {
if(al.HasTag("RG") &&
al.IsMapped() ){
string rgTag;
al.GetTag("RG",rgTag);
//cout<<rgTag<<endl;
if(rg2BamWriter.find(rgTag) == rg2BamWriter.end()){ //new: ignore completely
}else{
if( randomProb() <= rg2Fraction[ rgTag ] ){
rg2BamWriter[rgTag]->SaveAlignment(al);
//cout<<"wrote "<<rgTag<<endl;
} else{
//cout<<"skipped "<<rgTag<<endl;
}
}
}// else{
// string rgTag="unknown";
// //cout<<rgTag<<endl;
// if(rg2BamWriter.find(rgTag) == rg2BamWriter.end()){ //new
// cerr<<"Found new RG "<<rgTag<<endl;
// rg2BamWriter[rgTag] = new BamWriter();
// if ( !rg2BamWriter[rgTag]->Open(bamDirOutPrefix+"."+rgTag+".bam",header,references) ) {
// cerr << "Could not open output BAM file "<< bamDirOutPrefix<<"."<<rgTag<<".bam" << endl;
// return 1;
// }
// rg2BamWriter[rgTag]->SaveAlignment(al);
// }else{
// rg2BamWriter[rgTag]->SaveAlignment(al);
// }
// // cerr << "Cannot get RG tag for " << al.Name<<endl;
// // return 1;
// }
total++;
} //while al
reader.Close();
// writer.Close();
// unmapped.Close();
map<string,BamWriter *>::iterator rg2BamWriterIt;
for (rg2BamWriterIt =rg2BamWriter.begin();
rg2BamWriterIt!=rg2BamWriter.end();
rg2BamWriterIt++){
rg2BamWriterIt->second->Close();
}
cerr<<"Wrote succesfully "<<total<<" reads"<<endl;
return 0;
}
void GenericIndividualSnpCall::simpleSnpCall(string &fastaObj, BamReader &bamObj, int chrID, int leftPosition, int rightPosition, vector<Allele> &variantCandidates, map<int,list<tuple<char,int,int,double>>> &bamData)
{
set<int> BlockSnpPositions;
vector<Allele> BlockSnpAlleles;
// rewind
bamObj.Rewind();
// set region
bamObj.SetRegion(chrID, leftPosition, chrID, rightPosition);
BamAlignment al;
// search SNP positions in the region
while (bamObj.GetNextAlignment(al))
{
if (!GenericBamAlignmentTools::goodAlignment(al))
continue;
if (!al.HasTag("MD"))
continue;
vector<long> SnpInAlignment;
GenericBamAlignmentTools::getBamAlignmentMismatches(al, SnpInAlignment);
for (int i=0; i<SnpInAlignment.size(); i++)
{
BlockSnpPositions.insert(SnpInAlignment[i]);
}
}
// pileup visitor
SimpleSnpCallPileupVisitor visitor(&fastaObj, chrID, leftPosition, rightPosition, m_downSample, &BlockSnpPositions, &BlockSnpAlleles, &bamData);
PileupEngine SimpleSnpCallPileupEngine;
SimpleSnpCallPileupEngine.AddVisitor(&visitor);
// rewind
bamObj.Rewind();
// set region
bamObj.SetRegion(chrID, leftPosition, chrID, rightPosition);
// load data
while(bamObj.GetNextAlignment(al))
{
if (!GenericBamAlignmentTools::goodAlignment(al))
continue;
if (!GenericBamAlignmentTools::validMapQuality(al, m_minMapQuality))
continue;
if (!GenericBamAlignmentTools::validReadIdentity(al, m_maxMismatchFrac))
continue;
if (!GenericBamAlignmentTools::validReadLength(al, m_minReadLength))
continue;
if (!al.HasTag("MD"))
continue;
SimpleSnpCallPileupEngine.AddAlignment(al);
}
SimpleSnpCallPileupEngine.Flush();
// Filter SNP candidiate
for (int i=0; i<BlockSnpAlleles.size(); i++)
{
Allele allele = BlockSnpAlleles[i];
if (allele.m_alleleDepth < m_minSnpRead)
continue;
if (allele.m_alleleDepth < m_minSnpFrac*allele.m_globalDepth)
continue;
variantCandidates.push_back(allele);
}
}
int main (int argc, char *argv[]) {
if( (argc== 1) ||
(argc== 2 && string(argv[1]) == "-h") ||
(argc== 2 && string(argv[1]) == "-help") ||
(argc== 2 && string(argv[1]) == "--help") ){
cout<<"Usage:"<<argv[0]<<"<options> [in bam] [ref original length] [extension length]"<<endl;
cout<<"This program returns the same BAM file except with the XA flag for circular references"<<endl;
cout<<"Options:"<<endl;
//cout<<"\t-m\t\t\t\tUse mapped reads only"<<endl;
return 1;
}
for(int i=1;i<(argc-3);i++){ //all but the last 3 args
// if(string(argv[i]) == "-m" ){
// onlyMapped=true;
// continue;
// }
cerr<<"Unknown option "<<argv[i] <<" exiting"<<endl;
return 1;
}
string bamfiletopen = string( argv[argc-3]);
int origLength = destringify<int>(argv[argc-2]);
int extLength = destringify<int>(argv[argc-1]);
string outputFilename = "/dev/stdout";
BamReader reader;
if ( !reader.Open(bamfiletopen) ) {
cerr << "Could not open input BAM files." << endl;
return 1;
}
const SamHeader header = reader.GetHeader();
const RefVector references = reader.GetReferenceData();
BamWriter writer;
if ( !writer.Open(outputFilename, header, references) ) {
cerr << "Could not open output BAM file" << endl;
return 1;
}
BamAlignment al;
string nameTAG="XA";
while ( reader.GetNextAlignment(al) ) {
if(al.HasTag(nameTAG)) {
cerr << "ERROR: Read "<<al.Name<<" already has XA tags" << endl;
return 1;
}
writer.SaveAlignment(al);
} //while al
reader.Close();
writer.Close();
return 0;
}
