X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=splitmatrix.cpp;h=dabcd0469461f25a79ac707cf0e785c19974cc1e;hb=8dd3c225255d7084e3aff8740aa4f1f1cabb367a;hp=cd05cc46c172122ab9ee23490d38eda5e7308c22;hpb=cad05a21b084833b07808c1586e755be48fe7e1a;p=mothur.git diff --git a/splitmatrix.cpp b/splitmatrix.cpp index cd05cc4..dabcd04 100644 --- a/splitmatrix.cpp +++ b/splitmatrix.cpp @@ -9,9 +9,8 @@ #include "splitmatrix.h" #include "phylotree.h" -#include "sequencedb.h" -#include "onegapdist.h" -#include "dist.h" +#include "distancecommand.h" +#include "seqsummarycommand.h" /***********************************************************************/ @@ -26,12 +25,16 @@ SplitMatrix::SplitMatrix(string distfile, string name, string tax, float c, stri } /***********************************************************************/ -SplitMatrix::SplitMatrix(string ffile, string tax, float c, string t){ +SplitMatrix::SplitMatrix(string ffile, string name, string tax, float c, float cu, string t, int p, string output){ m = MothurOut::getInstance(); fastafile = ffile; + namefile = name; taxFile = tax; - cutoff = c; + cutoff = c; //tax level cutoff + distCutoff = cu; //for fasta method if you are creating distance matrix you need a cutoff for that method = t; + processors = p; + outputDir = output; } /***********************************************************************/ @@ -63,6 +66,8 @@ int SplitMatrix::splitDistance(){ if (large) { splitDistanceLarge(); } else { splitDistanceRAM(); } + + return 0; } catch(exception& e) { @@ -75,7 +80,7 @@ int SplitMatrix::splitDistance(){ int SplitMatrix::splitClassify(){ try { cutoff = int(cutoff); - + map seqGroup; map::iterator it; map::iterator it2; @@ -86,12 +91,12 @@ int SplitMatrix::splitClassify(){ PhyloTree* phylo = new PhyloTree(); ifstream in; - openInputFile(taxFile, in); + m->openInputFile(taxFile, in); //read in users taxonomy file and add sequences to tree string seqname, tax; while(!in.eof()){ - in >> seqname >> tax; gobble(in); + in >> seqname >> tax; m->gobble(in); phylo->addSeqToTree(seqname, tax); } in.close(); @@ -125,7 +130,6 @@ int SplitMatrix::splitClassify(){ createDistanceFilesFromTax(seqGroup, numGroups); } - return 0; } @@ -137,36 +141,128 @@ int SplitMatrix::splitClassify(){ /***********************************************************************/ int SplitMatrix::createDistanceFilesFromTax(map& seqGroup, int numGroups){ try { + map copyGroups = seqGroup; map::iterator it; - map::iterator it2; - map seqIndexInFasta; + set names; - //read fastafile - SequenceDB alignDB; - - ifstream filehandle; - openInputFile(fastafile, filehandle); - int numSeqs = 0; - while (!filehandle.eof()) { - //input sequence info into sequencedb - Sequence newSequence(filehandle); + for (int i = 0; i < numGroups; i++) { //remove old temp files, just in case + m->mothurRemove((fastafile + "." + toString(i) + ".temp")); + } - if (newSequence.getName() != "") { - alignDB.push_back(newSequence); - seqIndexInFasta[newSequence.getName()] = numSeqs; - numSeqs++; + ifstream in; + m->openInputFile(fastafile, in); + + //parse fastafile + ofstream outFile; + while (!in.eof()) { + Sequence query(in); m->gobble(in); + if (query.getName() != "") { + + it = seqGroup.find(query.getName()); + + //save names in case no namefile is given + if (namefile == "") { names.insert(query.getName()); } + + if (it != seqGroup.end()) { //not singleton + m->openOutputFileAppend((fastafile + "." + toString(it->second) + ".temp"), outFile); + query.printSequence(outFile); + outFile.close(); + + copyGroups.erase(query.getName()); + } } + } + in.close(); + + //warn about sequence in groups that are not in fasta file + for(it = copyGroups.begin(); it != copyGroups.end(); it++) { + m->mothurOut("ERROR: " + it->first + " is missing from your fastafile. This could happen if your taxonomy file is not unique and your fastafile is, or it could indicate and error."); m->mothurOutEndLine(); + exit(1); + } + + copyGroups.clear(); + + //process each distance file + for (int i = 0; i < numGroups; i++) { + + string options = "fasta=" + (fastafile + "." + toString(i) + ".temp") + ", processors=" + toString(processors) + ", cutoff=" + toString(distCutoff); + if (outputDir != "") { options += ", outputdir=" + outputDir; } + + Command* command = new DistanceCommand(options); - //takes care of white space - gobble(filehandle); + command->execute(); + delete command; + + m->mothurRemove((fastafile + "." + toString(i) + ".temp")); + + //remove old names files just in case + m->mothurRemove((namefile + "." + toString(i) + ".temp")); } - filehandle.close(); + + singleton = namefile + ".extra.temp"; + ofstream remainingNames; + m->openOutputFile(singleton, remainingNames); - Dist* distCalculator = new oneGapDist(); + bool wroteExtra = false; + + ifstream bigNameFile; + m->openInputFile(namefile, bigNameFile); + string name, nameList; + while(!bigNameFile.eof()){ + bigNameFile >> name >> nameList; m->gobble(bigNameFile); + + //did this sequence get assigned a group + it = seqGroup.find(name); + + if (it != seqGroup.end()) { + m->openOutputFileAppend((namefile + "." + toString(it->second) + ".temp"), outFile); + outFile << name << '\t' << nameList << endl; + outFile.close(); + }else{ + wroteExtra = true; + remainingNames << name << '\t' << nameList << endl; + } + } + bigNameFile.close(); -//still not done.... + for(int i=0;ihasPath(fastafile); } + string tempDistFile = outputDir + m->getRootName(m->getSimpleName((fastafile + "." + toString(i) + ".temp"))) + "dist"; + + //if there are valid distances + ifstream fileHandle; + fileHandle.open(tempDistFile.c_str()); + if(fileHandle) { + m->gobble(fileHandle); + if (!fileHandle.eof()) { //check for blank file - this could occur if all dists in group are above cutoff + map temp; + temp[tempDistFile] = tempNameFile; + dists.push_back(temp); + }else { + ifstream in; + m->openInputFile(tempNameFile, in); + + while(!in.eof()) { + in >> name >> nameList; m->gobble(in); + wroteExtra = true; + remainingNames << name << '\t' << nameList << endl; + } + in.close(); + m->mothurRemove(tempNameFile); + } + } + fileHandle.close(); + } + remainingNames.close(); + if (!wroteExtra) { + m->mothurRemove(singleton); + singleton = "none"; + } + + if (m->control_pressed) { for (int i = 0; i < dists.size(); i++) { m->mothurRemove((dists[i].begin()->first)); m->mothurRemove((dists[i].begin()->second)); } dists.clear(); } return 0; } @@ -182,11 +278,11 @@ int SplitMatrix::splitDistanceFileByTax(map& seqGroup, int numGroup map::iterator it2; ifstream dFile; - openInputFile(distFile, dFile); + m->openInputFile(distFile, dFile); ofstream outFile; for (int i = 0; i < numGroups; i++) { //remove old temp files, just in case - remove((distFile + "." + toString(i) + ".temp").c_str()); + m->mothurRemove((distFile + "." + toString(i) + ".temp")); } //for buffering the io to improve speed @@ -203,9 +299,9 @@ int SplitMatrix::splitDistanceFileByTax(map& seqGroup, int numGroup string seqA, seqB; float dist; - if (m->control_pressed) { dFile.close(); for (int i = 0; i < numGroups; i++) { remove((distFile + "." + toString(i) + ".temp").c_str()); } } + if (m->control_pressed) { dFile.close(); for (int i = 0; i < numGroups; i++) { m->mothurRemove((distFile + "." + toString(i) + ".temp")); } } - dFile >> seqA >> seqB >> dist; gobble(dFile); + dFile >> seqA >> seqB >> dist; m->gobble(dFile); //if both sequences are in the same group then they are within the cutoff it = seqGroup.find(seqA); @@ -214,7 +310,7 @@ int SplitMatrix::splitDistanceFileByTax(map& seqGroup, int numGroup if ((it != seqGroup.end()) && (it2 != seqGroup.end())) { //they are both not singletons if (it->second == it2->second) { //they are from the same group so add the distance if (numOutputs[it->second] > 30) { - openOutputFileAppend((distFile + "." + toString(it->second) + ".temp"), outFile); + m->openOutputFileAppend((distFile + "." + toString(it->second) + ".temp"), outFile); outFile << outputs[it->second] << seqA << '\t' << seqB << '\t' << dist << endl; outFile.close(); outputs[it->second] = ""; @@ -230,11 +326,11 @@ int SplitMatrix::splitDistanceFileByTax(map& seqGroup, int numGroup dFile.close(); for (int i = 0; i < numGroups; i++) { //remove old temp files, just in case - remove((namefile + "." + toString(i) + ".temp").c_str()); + m->mothurRemove((namefile + "." + toString(i) + ".temp")); //write out any remaining buffers if (numOutputs[i] > 0) { - openOutputFileAppend((distFile + "." + toString(i) + ".temp"), outFile); + m->openOutputFileAppend((distFile + "." + toString(i) + ".temp"), outFile); outFile << outputs[i]; outFile.close(); outputs[i] = ""; @@ -244,23 +340,23 @@ int SplitMatrix::splitDistanceFileByTax(map& seqGroup, int numGroup } ifstream bigNameFile; - openInputFile(namefile, bigNameFile); + m->openInputFile(namefile, bigNameFile); singleton = namefile + ".extra.temp"; ofstream remainingNames; - openOutputFile(singleton, remainingNames); + m->openOutputFile(singleton, remainingNames); bool wroteExtra = false; string name, nameList; while(!bigNameFile.eof()){ - bigNameFile >> name >> nameList; gobble(bigNameFile); + bigNameFile >> name >> nameList; m->gobble(bigNameFile); //did this sequence get assigned a group it = seqGroup.find(name); if (it != seqGroup.end()) { - openOutputFileAppend((namefile + "." + toString(it->second) + ".temp"), outFile); + m->openOutputFileAppend((namefile + "." + toString(it->second) + ".temp"), outFile); outFile << name << '\t' << nameList << endl; outFile.close(); }else{ @@ -269,29 +365,41 @@ int SplitMatrix::splitDistanceFileByTax(map& seqGroup, int numGroup } } bigNameFile.close(); - remainingNames.close(); - - if (!wroteExtra) { - remove(singleton.c_str()); - singleton = "none"; - } - + for(int i=0;i temp; temp[tempDistFile] = tempNameFile; dists.push_back(temp); + }else{ + ifstream in; + m->openInputFile(tempNameFile, in); + + while(!in.eof()) { + in >> name >> nameList; m->gobble(in); + wroteExtra = true; + remainingNames << name << '\t' << nameList << endl; + } + in.close(); + m->mothurRemove(tempNameFile); } } + remainingNames.close(); + + if (!wroteExtra) { + m->mothurRemove(singleton); + singleton = "none"; + } + if (m->control_pressed) { for (int i = 0; i < dists.size(); i++) { - remove((dists[i].begin()->first).c_str()); - remove((dists[i].begin()->second).c_str()); + m->mothurRemove((dists[i].begin()->first)); + m->mothurRemove((dists[i].begin()->second)); } dists.clear(); } @@ -318,7 +426,7 @@ int SplitMatrix::splitDistanceLarge(){ ofstream outFile; ifstream dFile; - openInputFile(distFile, dFile); + m->openInputFile(distFile, dFile); while(dFile){ string seqA, seqB; @@ -326,7 +434,7 @@ int SplitMatrix::splitDistanceLarge(){ dFile >> seqA >> seqB >> dist; - if (m->control_pressed) { dFile.close(); for(int i=0;i 0){ remove((distFile + "." + toString(i) + ".temp").c_str()); } } return 0; } + if (m->control_pressed) { dFile.close(); for(int i=0;i 0){ m->mothurRemove((distFile + "." + toString(i) + ".temp")); } } return 0; } if(dist < cutoff){ //cout << "in cutoff: " << dist << endl; @@ -454,7 +562,7 @@ int SplitMatrix::splitDistanceLarge(){ delete memblock; fileB.close(); - remove(fileName2.c_str()); + m->mothurRemove(fileName2); //write out the merged memory if (numOutputs[groupID] > 60) { @@ -514,7 +622,7 @@ int SplitMatrix::splitDistanceLarge(){ delete memblock; fileB.close(); - remove(fileName2.c_str()); + m->mothurRemove(fileName2); //write out the merged memory if (numOutputs[groupID] > 60) { @@ -532,7 +640,7 @@ int SplitMatrix::splitDistanceLarge(){ } } } - gobble(dFile); + m->gobble(dFile); } dFile.close(); @@ -570,7 +678,7 @@ int SplitMatrix::splitNames(vector >& groups){ while(bigNameFile){ bigNameFile >> name >> nameList; nameMap[name] = nameList; - gobble(bigNameFile); + m->gobble(bigNameFile); } bigNameFile.close(); @@ -617,8 +725,8 @@ int SplitMatrix::splitNames(vector >& groups){ if (m->control_pressed) { for (int i = 0; i < dists.size(); i++) { - remove((dists[i].begin()->first).c_str()); - remove((dists[i].begin()->second).c_str()); + m->mothurRemove((dists[i].begin()->first)); + m->mothurRemove((dists[i].begin()->second)); } dists.clear(); } @@ -639,7 +747,7 @@ int SplitMatrix::splitDistanceRAM(){ int numGroups = 0; ifstream dFile; - openInputFile(distFile, dFile); + m->openInputFile(distFile, dFile); while(dFile){ string seqA, seqB; @@ -647,7 +755,7 @@ int SplitMatrix::splitDistanceRAM(){ dFile >> seqA >> seqB >> dist; - if (m->control_pressed) { dFile.close(); for(int i=0;i 0){ remove((distFile + "." + toString(i) + ".temp").c_str()); } } return 0; } + if (m->control_pressed) { dFile.close(); for(int i=0;i 0){ m->mothurRemove((distFile + "." + toString(i) + ".temp")); } } return 0; } if(dist < cutoff){ //cout << "in cutoff: " << dist << endl; @@ -722,7 +830,7 @@ int SplitMatrix::splitDistanceRAM(){ } } } - gobble(dFile); + m->gobble(dFile); } dFile.close();