void SeedingData::writeCheckpoints(){
/* write the Seeds checkpoint */
if(m_parameters->writeCheckpoints() && !m_parameters->hasCheckpoint("SimpleSeeds")){
ofstream f(m_parameters->getCheckpointFile("SimpleSeeds").c_str());
ostringstream buffer;
cout<<"Rank "<<m_parameters->getRank()<<" is writing checkpoint SimpleSeeds"<<endl;
vector<GraphPath> * seeds = & m_SEEDING_seeds;
int count=(*seeds).size();
buffer.write((char*)&count, sizeof(int));
for(int i=0;i<(int)(*seeds).size();i++){
int length=(*seeds)[i].size();
buffer.write((char*)&length, sizeof(int));
for(int j=0;j<(int)(*seeds)[i].size();j++){
Kmer theKmer;
(*seeds)[i].at(j,&theKmer);
theKmer.write(&buffer);
CoverageDepth coverageValue=0;
coverageValue=(*seeds)[i].getCoverageAt(j);
buffer.write((char*)&coverageValue, sizeof(CoverageDepth));
flushFileOperationBuffer(false, &buffer, &f, CONFIG_FILE_IO_BUFFER_SIZE);
}
}
flushFileOperationBuffer(true, &buffer, &f, CONFIG_FILE_IO_BUFFER_SIZE);
f.close();
}
}
void GenomeNeighbourhood::processFinalList(){
/* we have a list of pairs
* there are duplicates
* and the list is still unfiltered.
*
* the first step is to select the best entry for
* any ordered pair where (a,b) and (b,a) are
* different
*/
/* now, select one entry for each */
/* now, select one entry for each */
/* cases:
*
*
VALID
1.
* *
-----------> ------------>
2.
* *
-----------> <------------
3.
* *
<------------ ------------->
4.
* *
<------------ <-------------
all other cases are invalid.
*
*/
ostringstream relations;
relations<<m_parameters->getPrefix()<<"/NeighbourhoodRelations.txt";
string file=relations.str();
ofstream f(file.c_str());
ostringstream operationBuffer;
operationBuffer<<"#LeftContigPath LengthInKmers DNAStrand PositionOnStrand";
operationBuffer<<" RightContigPath LengthInKmers DNAStrand PositionOnStrand";
operationBuffer<<" DistanceInKmers QualityControlStatus"<<endl;
for(int i=0;i<(int)m_finalList.size();i++){
PathHandle contig1=m_finalList[i].getContig1();
PathHandle contig2=m_finalList[i].getContig2();
int length1=m_contigLengths->operator[](contig1);
int length2=m_contigLengths->operator[](contig2);
Strand strand1=m_finalList[i].getStrand1();
Strand strand2=m_finalList[i].getStrand2();
int progression1=m_finalList[i].getProgression1();
int progression2=m_finalList[i].getProgression2();
int depth=m_finalList[i].getDepth();
bool valid=true;
int windows=(0x00000001 << 0x00000002);
int width1=length1/windows;
int width2=length2/windows;
/*The pair is considered valid unless the similarity is after the first 1/4 and before the last 3/4*/
if((progression1>width1 && progression1<(length1-width1-1)) || (progression2>width2 && progression2<(length2-width2-1))){
valid=false;
}
operationBuffer<<"contig-"<<contig1<<" "<<length1<<" "<<strand1<<" "<<progression1<<"";
operationBuffer<<" contig-"<<contig2<<" "<<length2<<" "<<strand2<<" "<<progression2<<"";
operationBuffer<<" "<<depth<<" ";
if(valid){
operationBuffer<<"PASS";
}else{
operationBuffer<<"FAIL";
}
operationBuffer<<endl;
flushFileOperationBuffer(false,&operationBuffer,&f,CONFIG_FILE_IO_BUFFER_SIZE);
}
flushFileOperationBuffer(true,&operationBuffer,&f,CONFIG_FILE_IO_BUFFER_SIZE);
f.close();
}
void GeneOntology::writeOntologyFiles(){
ostringstream theFile;
theFile<<m_parameters->getPrefix()<<"/BiologicalAbundances/_GeneOntology";
string directory=theFile.str();
createDirectory(directory.c_str());
theFile<<"/Terms";
ostringstream xmlFileStream;
xmlFileStream<<theFile.str()<<".xml";
ostringstream tsvFileStream;
tsvFileStream<<theFile.str()<<".tsv";
string xmlFile=xmlFileStream.str();
string tsvFile=tsvFileStream.str();
if(!m_gotGeneOntologyParameter){
return;
}
map<GeneOntologyIdentifier,int> modeCoverages;
map<GeneOntologyIdentifier,double> meanCoverages;
map<GeneOntologyIdentifier,double> estimatedProportions;
ofstream xmlStream(xmlFile.c_str());
ostringstream operationBuffer; //-------------
cout<<"TOTAL: "<<m_kmerObservationsWithGeneOntologies<<endl;
operationBuffer<<"<?xml version=\"1.0\" encoding=\"UTF-8\"?>"<<endl;
operationBuffer<<"<root>"<<endl;
LargeCount totalForTheGraph=m_searcher->getTotalNumberOfColoredKmerObservationsForANameSpace(COLOR_NAMESPACE_EMBL_CDS);
operationBuffer<<"<totalColoredKmerObservations>";
operationBuffer<<totalForTheGraph<<"</totalColoredKmerObservations>"<<endl;
// declare tsv files
map<string,FILE*> tsvFiles;
map<string,ostringstream*> tsvBuffers;
for(map<GeneOntologyIdentifier,map<CoverageDepth,int> >::iterator i=
m_ontologyTermFrequencies.begin();i!=m_ontologyTermFrequencies.end();i++){
GeneOntologyIdentifier handle=i->first;
int mode=0;
int modeCount=0;
int total=0;
LargeCount totalObservations=0;
for(map<CoverageDepth,int>::iterator j=i->second.begin();j!=i->second.end();j++){
CoverageDepth coverage=j->first;
int frequency=j->second;
if(frequency>modeCount){
mode=coverage;
modeCount=frequency;
}
total+=frequency;
totalObservations+=coverage*frequency;
}
double mean=totalObservations;
if(total!=0){
mean/=total;
}
#ifdef ASSERT
assert(modeCoverages.count(handle)==0);
assert(meanCoverages.count(handle)==0);
#endif /**/
modeCoverages[handle]=mode;
meanCoverages[handle]=mean; /**/
GeneOntologyDomain domain=getDomain(handle);
string domainName=getDomainName(domain);
operationBuffer<<"<geneOntologyTerm>"<<endl;
operationBuffer<<"<identifier>";
operationBuffer<<getGeneOntologyIdentifier(handle)<<"</identifier><name>";
operationBuffer<<getGeneOntologyName(handle)<<"</name>"<<endl;
operationBuffer<<"<domain>"<<domainName<<"</domain>"<<endl;
/* print paths to root */
printPathsFromRoot(handle,&operationBuffer);
operationBuffer<<"<modeKmerCoverage>"<<mode<<"</modeKmerCoverage>";
operationBuffer<<"<meanKmerCoverage>"<<mean<<"</meanKmerCoverage>"<<endl;
operationBuffer<<"<totalColoredKmerObservations>"<<totalObservations<<"</totalColoredKmerObservations>"<<endl;
double estimatedProportion=(0.0+totalObservations);
if(totalForTheGraph!=0){
estimatedProportion/=totalForTheGraph;
}
estimatedProportions[handle]=estimatedProportion;
m_termCounts[handle]=totalObservations;
operationBuffer<<"<proportion>"<<estimatedProportion<<"</proportion>"<<endl;
operationBuffer<<"<distribution>"<<endl;
operationBuffer<<"#Coverage Frequency"<<endl;
for(map<CoverageDepth,int>::iterator j=i->second.begin();j!=i->second.end();j++){
CoverageDepth coverage=j->first;
int frequency=j->second;
operationBuffer<<coverage<<" "<<frequency<<endl;
}
operationBuffer<<"</distribution></geneOntologyTerm>"<<endl;
flushFileOperationBuffer(false,&operationBuffer,&xmlStream,CONFIG_FILE_IO_BUFFER_SIZE);
// also output beautiful tsv file too
if(tsvFiles.count(domainName)==0){
ostringstream theFile;
theFile<<m_parameters->getPrefix()<<"/BiologicalAbundances/";
theFile<<"0.Profile.GeneOntologyDomain="<<domainName<<".tsv";
string tsvFile=theFile.str();
tsvFiles[domainName]=fopen(tsvFile.c_str(),"a");
tsvBuffers[domainName]=new ostringstream();
*(tsvBuffers[domainName])<<"#TermIdentifier TermName TermDomain TermProportion"<<endl;
}
*(tsvBuffers[domainName])<<getGeneOntologyIdentifier(handle)<<" ";
*(tsvBuffers[domainName])<<getGeneOntologyName(handle)<<" ";
*(tsvBuffers[domainName])<<domainName;
*(tsvBuffers[domainName])<<" "<<estimatedProportion<<endl;
}
operationBuffer<<"</root>"<<endl;
flushFileOperationBuffer(true,&operationBuffer,&xmlStream,CONFIG_FILE_IO_BUFFER_SIZE);
xmlStream.close();
ofstream tsvStream(tsvFile.c_str());
operationBuffer<<"#Identifier Name Mode k-mer coverage Mean k-mer coverage Proportion"<<endl;
// close tsv files
for(map<string,FILE*>::iterator i=tsvFiles.begin();i!=tsvFiles.end();i++){
string category=i->first;
FILE*file=i->second;
string text=tsvBuffers[category]->str();
fprintf(file,"%s",text.c_str());
delete tsvBuffers[category];
fclose(file);
}
tsvBuffers.clear();
tsvFiles.clear();
for(map<GeneOntologyIdentifier,map<CoverageDepth,int> >::iterator i=
m_ontologyTermFrequencies.begin();i!=m_ontologyTermFrequencies.end();i++){
GeneOntologyIdentifier handle=i->first;
operationBuffer<<getGeneOntologyIdentifier(handle)<<" ";
operationBuffer<<getGeneOntologyName(handle)<<" ";
operationBuffer<<modeCoverages[handle]<<" ";
operationBuffer<<meanCoverages[handle]<<" ";
operationBuffer<<estimatedProportions[handle]<<endl;
flushFileOperationBuffer(false,&operationBuffer,&tsvStream,CONFIG_FILE_IO_BUFFER_SIZE);
}
flushFileOperationBuffer(true,&operationBuffer,&tsvStream,CONFIG_FILE_IO_BUFFER_SIZE);
tsvStream.close();
}
void GeneOntology::writeOntologyProfile(GeneOntologyDomain domain){
int maximumDepth=getDomainDepth(domain);
cout<<"[GeneOntology] maximum depth for GeneOntologyDomain "<<domain<<" is "<<maximumDepth<<endl;
string domainName="NULL";
if(domain==GENE_ONTOLOGY_DOMAIN_biological_process){
domainName=GENE_ONTOLOGY_DOMAIN_biological_process_STRING;
}else if(domain==GENE_ONTOLOGY_DOMAIN_cellular_component){
domainName=GENE_ONTOLOGY_DOMAIN_cellular_component_STRING;
}else if(domain==GENE_ONTOLOGY_DOMAIN_molecular_function){
domainName=GENE_ONTOLOGY_DOMAIN_molecular_function_STRING;
}
#ifdef ASSERT
assert(domainName!="NULL");
#endif
LargeCount totalForTheGraph=m_searcher->getTotalNumberOfColoredKmerObservationsForANameSpace(COLOR_NAMESPACE_EMBL_CDS);
for(int depth=0;depth<maximumDepth;depth++){
// create the file for the domain and given depth.
ostringstream operationBuffer;
ostringstream fileName;
fileName<<m_parameters->getPrefix()<<"/BiologicalAbundances/_GeneOntology";
fileName<<"/"<<domainName<<".Depth="<<depth<<".tsv";
string file2=fileName.str();
ofstream file;
for(map<GeneOntologyIdentifier,int>::iterator i=m_recursiveCounts.begin();
i!=m_recursiveCounts.end();i++){
GeneOntologyIdentifier handle=i->first;
int count=i->second;
if(count==0){
continue;
}
if(!hasDepth(handle)){
continue;
}
if(getGeneOntologyDepth(handle)!=depth){
continue;
}
if(getDomain(handle)!=domain){
continue;
}
double proportion=count;
if(totalForTheGraph!=0){
proportion/=totalForTheGraph;
}
if(!file.is_open()){
file.open(file2.c_str());
operationBuffer<<"#Identifier Name Proportion Observations Total"<<endl;
}
operationBuffer<<getGeneOntologyIdentifier(handle);
operationBuffer<<" "<<getGeneOntologyName(handle)<<" ";
operationBuffer<<proportion;
operationBuffer<<" "<<count<<" "<<totalForTheGraph<<endl;
flushFileOperationBuffer(false,&operationBuffer,&file,CONFIG_FILE_IO_BUFFER_SIZE);
}
if(file.is_open()){
flushFileOperationBuffer(true,&operationBuffer,&file,CONFIG_FILE_IO_BUFFER_SIZE);
file.close();
}
}
}
void PhylogenyViewer::showObservations_XML(ostream*stream){
ostringstream operationBuffer;
/* build a mashup for the ranks
* this will contain total at each level */
map<string,LargeCount> rankRecursiveObservations;
map<string,LargeCount> rankSelfObservations;
populateRanks(&rankSelfObservations,&rankRecursiveObservations);
operationBuffer<<"<?xml version=\"1.0\" encoding=\"UTF-8\"?>"<<endl;
operationBuffer<<"<root>"<<endl;
/* add the sample name in the XML file */
operationBuffer<<"<sample>";
operationBuffer<<m_parameters->getSampleName();
operationBuffer<<"</sample>"<<endl;
operationBuffer<<"<totalAssembledKmerObservations>"<<m_totalNumberOfKmerObservations<<"</totalAssembledKmerObservations>"<<endl;
LargeCount totalColoredAssembledKmerObservations=m_totalNumberOfKmerObservations-m_unknown;
operationBuffer<<"<totalColoredAssembledKmerObservations>"<<totalColoredAssembledKmerObservations<<"</totalColoredAssembledKmerObservations>"<<endl;
operationBuffer<<"<ranks>"<<endl;
for(map<string,LargeCount>::iterator i=rankSelfObservations.begin();i!=rankSelfObservations.end();i++){
string rank=i->first;
#ifdef ASSERT
assert(rankRecursiveObservations.count(rank)>0);
assert(rankSelfObservations.count(rank)>0);
#endif
operationBuffer<<"<entry><rank>"<<rank<<"</rank><self><kmerObservations>";
operationBuffer<<rankSelfObservations[rank]<<"</kmerObservations></self>";
operationBuffer<<"<recursive><kmerObservations>"<<rankRecursiveObservations[rank];
operationBuffer<<"</kmerObservations></recursive></entry>"<<endl;
}
operationBuffer<<"</ranks>"<<endl;
operationBuffer<<"<entry>";
operationBuffer<<"<taxon><identifier>unknown</identifier><name>unknown</name><rank>unknown</rank></taxon>"<<endl;
operationBuffer<<"<path></path>"<<endl;
operationBuffer<<"<self><kmerObservations>"<<m_unknown<<"</kmerObservations>";
double ratio=m_unknown;
if(m_totalNumberOfKmerObservations!=0)
ratio/=m_totalNumberOfKmerObservations;
operationBuffer<<"<proportion>"<<ratio<<"</proportion>";
operationBuffer<<"<coloredProportion>0</coloredProportion>";
operationBuffer<<"<coloredProportionInRank>0</coloredProportionInRank></self></entry>"<<endl;
// declare tsv files
map<string,FILE*> tsvFiles;
map<string,ostringstream*> tsvBuffers;
for(map<TaxonIdentifier,string>::iterator i=m_taxonNames.begin();
i!=m_taxonNames.end();i++){
TaxonIdentifier taxon=i->first;
LargeCount count=getSelfCount(taxon);
string rank=getTaxonRank(taxon);
#ifdef ASSERT
assert(rankSelfObservations.count(rank)>0);
assert(rankRecursiveObservations.count(rank)>0);
#endif
LargeCount rankRecursiveCount=rankRecursiveObservations[rank];
#ifdef ASSERT
LargeCount rankSelfCount=rankSelfObservations[rank]; //-
assert(rankSelfCount>=0);
assert(rankRecursiveCount>=0);
#endif
LargeCount recursiveCount=getRecursiveCount(taxon);
if(recursiveCount==0){
continue;
}
operationBuffer<<"<entry>"<<endl;
printTaxon_XML(taxon,&operationBuffer);
vector<TaxonIdentifier> path;
getTaxonPathFromRoot(taxon,&path);
printTaxonPath_XML(taxon,&path,&operationBuffer);
operationBuffer<<"<self>"<<endl;
operationBuffer<<"<kmerObservations>"<<count<<"</kmerObservations>";
double ratio=count;
if(m_totalNumberOfKmerObservations!=0)
ratio/=m_totalNumberOfKmerObservations;
operationBuffer<<"<proportion>"<<ratio<<"</proportion>";
double coloredRatio=count;
if(totalColoredAssembledKmerObservations!=0){
coloredRatio/=totalColoredAssembledKmerObservations;
}
operationBuffer<<"<coloredProportion>"<<coloredRatio<<"</coloredProportion>";
operationBuffer<<"</self>"<<endl;
operationBuffer<<"<recursive>";
operationBuffer<<"<kmerObservations>";
operationBuffer<<recursiveCount;
operationBuffer<<"</kmerObservations>"<<endl;
double ratio2=recursiveCount;
if(m_totalNumberOfKmerObservations!=0)
ratio2/=m_totalNumberOfKmerObservations;
operationBuffer<<"<proportion>"<<ratio2<<"</proportion>";
double coloredRatio2=recursiveCount;
if(totalColoredAssembledKmerObservations!=0){
coloredRatio2/=totalColoredAssembledKmerObservations;
}
operationBuffer<<"<coloredProportion>"<<coloredRatio2<<"</coloredProportion>";
double coloredRatioInRank=recursiveCount;
if(rankRecursiveCount!=0){
coloredRatioInRank/=rankRecursiveCount;
}
operationBuffer<<"<coloredProportionInRank>"<<coloredRatioInRank<<"</coloredProportionInRank>";
operationBuffer<<"</recursive>"<<endl;
operationBuffer<<"</entry>"<<endl;
flushFileOperationBuffer(false,&operationBuffer,stream,CONFIG_FILE_IO_BUFFER_SIZE);
// add data to the tsv file
if(tsvFiles.count(rank)==0){
ostringstream theFile;
theFile<<m_parameters->getPrefix()<<"/BiologicalAbundances/";
theFile<<"0.Profile.TaxonomyRank="<<rank<<".tsv";
string tsvFile=theFile.str();
tsvFiles[rank]=fopen(tsvFile.c_str(),"a");
tsvBuffers[rank]=new ostringstream();
*(tsvBuffers[rank])<<"#TaxonIdentifier TaxonName TaxonRank TaxonProportion"<<endl;
}
string name=getTaxonName(taxon);
*(tsvBuffers[rank])<<taxon<<" "<<name<<" "<<rank;
*(tsvBuffers[rank])<<" "<<coloredRatioInRank<<endl;
}
// close XML files
operationBuffer<<"</root>"<<endl;
flushFileOperationBuffer(true,&operationBuffer,stream,CONFIG_FILE_IO_BUFFER_SIZE);
// close tsv files
for(map<string,FILE*>::iterator i=tsvFiles.begin();i!=tsvFiles.end();i++){
string rank=i->first;
FILE*file=i->second;
string text=tsvBuffers[rank]->str();
fprintf(file,"%s",text.c_str());
delete tsvBuffers[rank];
fclose(file);
}
tsvBuffers.clear();
tsvFiles.clear();
}
