#include "trimseqscommand.h"
#include "needlemanoverlap.hpp"
+//**********************************************************************************************************************
+vector<string> TrimSeqsCommand::getValidParameters(){
+ try {
+ string Array[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "group","minlength", "maxlength", "qfile",
+ "qthreshold", "qwindowaverage", "qstepsize", "qwindowsize", "qaverage", "rollaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
+ vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "getValidParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+TrimSeqsCommand::TrimSeqsCommand(){
+ try {
+ abort = true;
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["fasta"] = tempOutNames;
+ outputTypes["qual"] = tempOutNames;
+ outputTypes["group"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "TrimSeqsCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> TrimSeqsCommand::getRequiredParameters(){
+ try {
+ string Array[] = {"fasta"};
+ vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "getRequiredParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> TrimSeqsCommand::getRequiredFiles(){
+ try {
+ vector<string> myArray;
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "getRequiredFiles");
+ exit(1);
+ }
+}
//***************************************************************************************************************
TrimSeqsCommand::TrimSeqsCommand(string option) {
else {
//valid paramters for this command
- string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
+ string AlignArray[] = {"fasta", "flip", "group","oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
"qthreshold", "qwindowaverage", "qstepsize", "qwindowsize", "qaverage", "rollaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["fasta"] = tempOutNames;
+ outputTypes["qual"] = tempOutNames;
+ outputTypes["group"] = tempOutNames;
+
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.validFile(parameters, "inputdir", false);
if (inputDir == "not found"){ inputDir = ""; }
it = parameters.find("fasta");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["fasta"] = inputDir + it->second; }
}
it = parameters.find("oligos");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["oligos"] = inputDir + it->second; }
}
it = parameters.find("qfile");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["qfile"] = inputDir + it->second; }
}
+
+ it = parameters.find("group");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["group"] = inputDir + it->second; }
+ }
}
//check for required parameters
fastaFile = validParameter.validFile(parameters, "fasta", true);
- if (fastaFile == "not found") { m->mothurOut("fasta is a required parameter for the screen.seqs command."); m->mothurOutEndLine(); abort = true; }
+ if (fastaFile == "not found") { m->mothurOut("fasta is a required parameter for the trim.seqs command."); m->mothurOutEndLine(); abort = true; }
else if (fastaFile == "not open") { abort = true; }
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
outputDir = "";
- outputDir += hasPath(fastaFile); //if user entered a file with a path then preserve it
+ outputDir += m->hasPath(fastaFile); //if user entered a file with a path then preserve it
}
string temp;
temp = validParameter.validFile(parameters, "flip", false);
if (temp == "not found"){ flip = 0; }
- else if(isTrue(temp)) { flip = 1; }
+ else if(m->isTrue(temp)) { flip = 1; }
temp = validParameter.validFile(parameters, "oligos", true);
if (temp == "not found"){ oligoFile = ""; }
else if(temp == "not open"){ abort = true; }
else { oligoFile = temp; }
+ temp = validParameter.validFile(parameters, "group", true);
+ if (temp == "not found"){ groupfile = ""; }
+ else if(temp == "not open"){ abort = true; }
+ else { groupfile = temp; }
+
temp = validParameter.validFile(parameters, "maxambig", false); if (temp == "not found") { temp = "-1"; }
convert(temp, maxAmbig);
convert(temp, qThreshold);
temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "F"; }
- qtrim = isTrue(temp);
+ qtrim = m->isTrue(temp);
temp = validParameter.validFile(parameters, "rollaverage", false); if (temp == "not found") { temp = "0"; }
convert(temp, qRollAverage);
convert(temp, qAverage);
temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; }
- allFiles = isTrue(temp);
+ allFiles = m->isTrue(temp);
temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
convert(temp, processors);
- if(allFiles && oligoFile == ""){
- m->mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); m->mothurOutEndLine();
+ if ((oligoFile != "") && (groupfile != "")) {
+ m->mothurOut("You given both a oligos file and a groupfile, only one is allowed."); m->mothurOutEndLine(); abort = true;
+ }
+
+
+ if(allFiles && (oligoFile == "") && (groupfile == "")){
+ m->mothurOut("You selected allfiles, but didn't enter an oligos or group file. Ignoring the allfiles request."); m->mothurOutEndLine();
}
if((qAverage != 0 && qThreshold != 0) && qFileName == ""){
m->mothurOut("You didn't provide a quality file name, quality criteria will be ignored."); m->mothurOutEndLine();
void TrimSeqsCommand::help(){
try {
m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
- m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
+ m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, group, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
m->mothurOut("The fasta parameter is required.\n");
+ m->mothurOut("The group parameter allows you to enter a group file for your fasta file.\n");
m->mothurOut("The flip parameter will output the reverse compliment of your trimmed sequence. The default is false.\n");
m->mothurOut("The oligos parameter .... The default is "".\n");
m->mothurOut("The maxambig parameter .... The default is -1.\n");
numFPrimers = 0; //this needs to be initialized
numRPrimers = 0;
-
- string trimSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.fasta";
- outputNames.push_back(trimSeqFile);
- string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
- outputNames.push_back(scrapSeqFile);
- string trimQualFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.qual";
- outputNames.push_back(trimQualFile);
- string scrapQualFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.qual";
- outputNames.push_back(scrapQualFile);
- string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
-
vector<string> fastaFileNames;
vector<string> qualFileNames;
+
+ string trimSeqFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "trim.fasta";
+ outputNames.push_back(trimSeqFile); outputTypes["fasta"].push_back(trimSeqFile);
+ string scrapSeqFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "scrap.fasta";
+ outputNames.push_back(scrapSeqFile); outputTypes["fasta"].push_back(scrapSeqFile);
+ string trimQualFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "trim.qual";
+ string scrapQualFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "scrap.qual";
+ if (qFileName != "") { outputNames.push_back(trimQualFile); outputNames.push_back(scrapQualFile); outputTypes["qual"].push_back(trimQualFile); outputTypes["qual"].push_back(scrapQualFile); }
+ string groupFile = "";
+ if (groupfile == "") { groupFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "groups"; }
+ else{
+ groupFile = outputDir + m->getRootName(m->getSimpleName(groupfile)) + "trim.groups";
+ outputNames.push_back(groupFile); outputTypes["group"].push_back(groupFile);
+ groupMap = new GroupMap(groupfile);
+ groupMap->readMap();
+
+ if(allFiles){
+ for (int i = 0; i < groupMap->namesOfGroups.size(); i++) {
+ groupToIndex[groupMap->namesOfGroups[i]] = i;
+ groupVector.push_back(groupMap->namesOfGroups[i]);
+ fastaFileNames.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + groupMap->namesOfGroups[i] + ".fasta"));
+
+ //we append later, so we want to clear file
+ ofstream outRemove;
+ m->openOutputFile(fastaFileNames[i], outRemove);
+ outRemove.close();
+ if(qFileName != ""){
+ qualFileNames.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + groupMap->namesOfGroups[i] + ".qual"));
+ ofstream outRemove2;
+ m->openOutputFile(qualFileNames[i], outRemove2);
+ outRemove2.close();
+ }
+ }
+ }
+ comboStarts = fastaFileNames.size()-1;
+ }
+
if(oligoFile != ""){
- outputNames.push_back(groupFile);
+ outputNames.push_back(groupFile); outputTypes["group"].push_back(groupFile);
getOligos(fastaFileNames, qualFileNames);
}
- if(qFileName != "") { setLines(qFileName, qLines); }
-
+ vector<unsigned long int> fastaFilePos;
+ vector<unsigned long int> qFilePos;
+
+ setLines(fastaFile, qFileName, fastaFilePos, qFilePos);
+
+ for (int i = 0; i < (fastaFilePos.size()-1); i++) {
+ lines.push_back(new linePair(fastaFilePos[i], fastaFilePos[(i+1)]));
+ if (qFileName != "") { qLines.push_back(new linePair(qFilePos[i], qFilePos[(i+1)])); }
+ }
+ if(qFileName == "") { qLines = lines; } //files with duds
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(processors == 1){
- ifstream inFASTA;
- int numSeqs;
- openInputFile(fastaFile, inFASTA);
- getNumSeqs(inFASTA, numSeqs);
- inFASTA.close();
-
- lines.push_back(new linePair(0, numSeqs));
-
- driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], lines[0]);
-
- for (int j = 0; j < fastaFileNames.size(); j++) {
- rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str());
- }
- if(qFileName != ""){
- for (int j = 0; j < qualFileNames.size(); j++) {
- rename((qualFileNames[j] + toString(getpid()) + ".temp").c_str(), qualFileNames[j].c_str());
- }
- }
-
+ driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], qLines[0]);
}else{
- setLines(fastaFile, lines);
- if(qFileName == "") { qLines = lines; }
-
createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames);
-
- rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str());
- rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str());
- rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str());
-
- if(qFileName != ""){
- rename((trimQualFile + toString(processIDS[0]) + ".temp").c_str(), trimQualFile.c_str());
- rename((scrapQualFile + toString(processIDS[0]) + ".temp").c_str(), scrapQualFile.c_str());
- }
-
-
- for (int j = 0; j < fastaFileNames.size(); j++) {
- rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str());
- }
- if(qFileName != ""){
- for (int j = 0; j < qualFileNames.size(); j++) {
- rename((qualFileNames[j] + toString(getpid()) + ".temp").c_str(), qualFileNames[j].c_str());
- }
- }
-
- //append files
- for(int i=1;i<processors;i++){
- appendFiles((trimSeqFile + toString(processIDS[i]) + ".temp"), trimSeqFile);
- remove((trimSeqFile + toString(processIDS[i]) + ".temp").c_str());
- appendFiles((scrapSeqFile + toString(processIDS[i]) + ".temp"), scrapSeqFile);
- remove((scrapSeqFile + toString(processIDS[i]) + ".temp").c_str());
-
- appendFiles((trimQualFile + toString(processIDS[i]) + ".temp"), trimQualFile);
- remove((trimQualFile + toString(processIDS[i]) + ".temp").c_str());
- appendFiles((scrapQualFile + toString(processIDS[i]) + ".temp"), scrapQualFile);
- remove((scrapQualFile + toString(processIDS[i]) + ".temp").c_str());
-
- appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
- remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
- for (int j = 0; j < fastaFileNames.size(); j++) {
- appendFiles((fastaFileNames[j] + toString(processIDS[i]) + ".temp"), fastaFileNames[j]);
- remove((fastaFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
- }
-
- if(qFileName != ""){
- for (int j = 0; j < qualFileNames.size(); j++) {
- appendFiles((qualFileNames[j] + toString(processIDS[i]) + ".temp"), qualFileNames[j]);
- remove((qualFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
- }
- }
-
-
- }
- }
-
- if (m->control_pressed) { return 0; }
+ }
#else
- ifstream inFASTA;
- int numSeqs;
- openInputFile(fastaFile, inFASTA);
- getNumSeqs(inFASTA, numSeqs);
- inFASTA.close();
-
- lines.push_back(new linePair(0, numSeqs));
-
- driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], lines[0]);
-
- for (int j = 0; j < fastaFileNames.size(); j++) {
- rename((fastaFileNames[j] + toString(j) + ".temp").c_str(), fastaFileNames[j].c_str());
- }
- if(qFileName != ""){
- for (int j = 0; j < qualFileNames.size(); j++) {
- rename((qualFileNames[j] + toString(j) + ".temp").c_str(), qualFileNames[j].c_str());
- }
- }
-
- if (m->control_pressed) { return 0; }
+ driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, trimQualFile, scrapQualFile, groupFile, fastaFileNames, qualFileNames, lines[0], qLines[0]);
#endif
-
+
+ if (m->control_pressed) { return 0; }
for(int i=0;i<fastaFileNames.size();i++){
- if (isBlank(fastaFileNames[i])) { remove(fastaFileNames[i].c_str()); }
- else if (filesToRemove.count(fastaFileNames[i]) > 0) { remove(fastaFileNames[i].c_str()); }
+
+ if (m->isBlank(fastaFileNames[i])) { remove(fastaFileNames[i].c_str()); }
+ else if (filesToRemove.count(fastaFileNames[i]) > 0) { remove(fastaFileNames[i].c_str()); }
else {
ifstream inFASTA;
string seqName;
- openInputFile(fastaFileNames[i], inFASTA);
+ m->openInputFile(fastaFileNames[i], inFASTA);
ofstream outGroups;
- string outGroupFilename = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "groups";
- openOutputFile(outGroupFilename, outGroups);
- outputNames.push_back(outGroupFilename);
+ string outGroupFilename = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[i])) + "groups";
+ m->openOutputFile(outGroupFilename, outGroups);
+ outputNames.push_back(outGroupFilename); outputTypes["group"].push_back(outGroupFilename);
string thisGroup = "";
if (i > comboStarts) {
if(itCombo->second == i){ thisGroup = itCombo->first; combos.erase(itCombo); break; }
}
}else{ thisGroup = groupVector[i]; }
-
+
while(!inFASTA.eof()){
if(inFASTA.get() == '>'){
inFASTA >> seqName;
if(qFileName != ""){
for(int i=0;i<qualFileNames.size();i++){
- if (isBlank(qualFileNames[i])) { remove(qualFileNames[i].c_str()); }
- else if (filesToRemove.count(qualFileNames[i]) > 0) { remove(qualFileNames[i].c_str()); }
+ if (m->isBlank(qualFileNames[i])) { remove(qualFileNames[i].c_str()); }
+ else if (filesToRemove.count(qualFileNames[i]) > 0) { remove(qualFileNames[i].c_str()); }
else {
ifstream inQual;
string seqName;
- openInputFile(qualFileNames[i], inQual);
+ m->openInputFile(qualFileNames[i], inQual);
// ofstream outGroups;
//
// string thisGroup = "";
try {
ofstream outFASTA;
- int able = openOutputFile(trimFile, outFASTA);
+ int able = m->openOutputFile(trimFile, outFASTA);
ofstream scrapFASTA;
- openOutputFile(scrapFile, scrapFASTA);
+ m->openOutputFile(scrapFile, scrapFASTA);
ofstream outQual;
ofstream scrapQual;
if(qFileName != ""){
- openOutputFile(trimQFile, outQual);
- openOutputFile(scrapQFile, scrapQual);
+ m->openOutputFile(trimQFile, outQual);
+ m->openOutputFile(scrapQFile, scrapQual);
}
ofstream outGroups;
- vector<ofstream*> fastaFileNames;
- vector<ofstream*> qualFileNames;
-
if (oligoFile != "") {
- openOutputFile(groupFile, outGroups);
- for (int i = 0; i < fastaNames.size(); i++) {
- #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
- if(qFileName != ""){
- qualFileNames.push_back(new ofstream((qualNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
- }
- #else
- fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(i) + ".temp").c_str(), ios::ate));
- if(qFileName != ""){
- qualFileNames.push_back(new ofstream((qualNames[i] + toString(i) + ".temp").c_str(), ios::ate));
- }
- #endif
- }
+ m->openOutputFile(groupFile, outGroups);
}
ifstream inFASTA;
- openInputFile(filename, inFASTA);
+ m->openInputFile(filename, inFASTA);
+ inFASTA.seekg(line->start);
ifstream qFile;
- if(qFileName != "") { openInputFile(qFileName, qFile); }
+ if(qFileName != "") { m->openInputFile(qFileName, qFile); qFile.seekg(qline->start); }
- qFile.seekg(qline->start);
- inFASTA.seekg(line->start);
- for(int i=0;i<line->num;i++){
+ for (int i = 0; i < fastaNames.size(); i++) { //clears old file
+ ofstream temp;
+ m->openOutputFile(fastaNames[i], temp);
+ temp.close();
+ }
+ for (int i = 0; i < qualNames.size(); i++) { //clears old file
+ ofstream temp;
+ m->openOutputFile(qualNames[i], temp);
+ temp.close();
+ }
+
+
+ bool done = false;
+ int count = 0;
+
+ while (!done) {
if (m->control_pressed) {
inFASTA.close(); outFASTA.close(); scrapFASTA.close();
if (oligoFile != "") { outGroups.close(); }
-
- for(int i=0;i<fastaFileNames.size();i++){ fastaFileNames[i]->close(); delete fastaFileNames[i]; }
if(qFileName != ""){
qFile.close();
- for(int i=0;i<qualFileNames.size();i++){ qualFileNames[i]->close(); delete qualFileNames[i]; }
}
for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
int success = 1;
- Sequence currSeq(inFASTA);
+
+ Sequence currSeq(inFASTA); m->gobble(inFASTA);
+
QualityScores currQual;
if(qFileName != ""){
- currQual = QualityScores(qFile, currSeq.getNumBases());
+ currQual = QualityScores(qFile, currSeq.getNumBases()); m->gobble(qFile);
}
string origSeq = currSeq.getUnaligned();
}
outGroups << currSeq.getName() << '\t' << thisGroup << endl;
if(allFiles){
- currSeq.printSequence(*fastaFileNames[indexToFastaFile]);
+ ofstream outTemp;
+ m->openOutputFileAppend(fastaNames[indexToFastaFile], outTemp);
+ //currSeq.printSequence(*fastaFileNames[indexToFastaFile]);
+ currSeq.printSequence(outTemp);
+ outTemp.close();
if(qFileName != ""){
- currQual.printQScores(*qualFileNames[indexToFastaFile]);
+ //currQual.printQScores(*qualFileNames[indexToFastaFile]);
+ ofstream outTemp2;
+ m->openOutputFileAppend(qualNames[indexToFastaFile], outTemp2);
+ currQual.printQScores(outTemp2);
+ outTemp2.close();
+ }
+ }
+ }
+
+ if (groupfile != "") {
+ string thisGroup = groupMap->getGroup(currSeq.getName());
+
+ if (thisGroup != "not found") {
+ outGroups << currSeq.getName() << '\t' << thisGroup << endl;
+ if (allFiles) {
+ ofstream outTemp;
+ m->openOutputFileAppend(fastaNames[groupToIndex[thisGroup]], outTemp);
+ currSeq.printSequence(outTemp);
+ outTemp.close();
+ if(qFileName != ""){
+ ofstream outTemp2;
+ m->openOutputFileAppend(qualNames[groupToIndex[thisGroup]], outTemp2);
+ currQual.printQScores(outTemp2);
+ outTemp2.close();
+ }
+ }
+ }else{
+ m->mothurOut(currSeq.getName() + " is not in your groupfile, adding to group XXX."); m->mothurOutEndLine();
+ outGroups << currSeq.getName() << '\t' << "XXX" << endl;
+ if (allFiles) {
+ m->mothurOut("[ERROR]: " + currSeq.getName() + " will not be added to any .group.fasta or .group.qual file."); m->mothurOutEndLine();
}
}
}
currSeq.printSequence(scrapFASTA);
currQual.printQScores(scrapQual);
}
+ count++;
}
- gobble(inFASTA);
- gobble(qFile);
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ unsigned long int pos = inFASTA.tellg();
+ if ((pos == -1) || (pos >= line->end)) { break; }
+ #else
+ if (inFASTA.eof()) { break; }
+ #endif
+
+ //report progress
+ if((count) % 1000 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); }
+
}
+ //report progress
+ if((count) % 1000 != 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); }
+
inFASTA.close();
outFASTA.close();
if (oligoFile != "") { outGroups.close(); }
if(qFileName != "") { qFile.close(); scrapQual.close(); outQual.close(); }
- for(int i=0;i<fastaFileNames.size();i++){
- fastaFileNames[i]->close();
- delete fastaFileNames[i];
- }
-
- if(qFileName != ""){
- for(int i=0;i<qualFileNames.size();i++){
- qualFileNames[i]->close();
- delete qualFileNames[i];
- }
- }
-
- return 0;
+ return count;
}
catch(exception& e) {
m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string trimQFile, string scrapQFile, string groupFile, vector<string> fastaNames, vector<string> qualNames) {
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- int process = 0;
+ int process = 1;
int exitCommand = 1;
processIDS.clear();
processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
+ for (int i = 0; i < fastaNames.size(); i++) {
+ fastaNames[i] = (fastaNames[i] + toString(getpid()) + ".temp");
+ //clear old file if it exists
+ ofstream temp;
+ m->openOutputFile(fastaNames[i], temp);
+ temp.close();
+ if(qFileName != ""){
+ qualNames[i] = (qualNames[i] + toString(getpid()) + ".temp");
+ //clear old file if it exists
+ ofstream temp2;
+ m->openOutputFile(qualNames[i], temp2);
+ temp2.close();
+ }
+ }
+
driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (trimQFile + toString(getpid()) + ".temp"), (scrapQFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, qualNames, lines[process], qLines[process]);
exit(0);
- }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+ }else {
+ m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
+ for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
+ exit(0);
+ }
}
+ //parent do my part
+ for (int i = 0; i < fastaNames.size(); i++) {
+ //clear old file if it exists
+ ofstream temp;
+ m->openOutputFile(fastaNames[i], temp);
+ temp.close();
+ if(qFileName != ""){
+ //clear old file if it exists
+ ofstream temp2;
+ m->openOutputFile(qualNames[i], temp2);
+ temp2.close();
+ }
+ }
+
+ driverCreateTrim(filename, qFileName, trimFile, scrapFile, trimQFile, scrapQFile, groupFile, fastaNames, qualNames, lines[0], qLines[0]);
+
+
//force parent to wait until all the processes are done
- for (int i=0;i<processors;i++) {
+ for (int i=0;i<processIDS.size();i++) {
int temp = processIDS[i];
wait(&temp);
}
+ //append files
+ for(int i=0;i<processIDS.size();i++){
+ m->mothurOut("Appending files from process " + processIDS[i]); m->mothurOutEndLine();
+
+ m->appendFiles((trimFile + toString(processIDS[i]) + ".temp"), trimFile);
+ remove((trimFile + toString(processIDS[i]) + ".temp").c_str());
+ m->appendFiles((scrapFile + toString(processIDS[i]) + ".temp"), scrapFile);
+ remove((scrapFile + toString(processIDS[i]) + ".temp").c_str());
+
+ m->mothurOut("Done with fasta files"); m->mothurOutEndLine();
+
+ if(qFileName != ""){
+ m->appendFiles((trimQFile + toString(processIDS[i]) + ".temp"), trimQFile);
+ remove((trimQFile + toString(processIDS[i]) + ".temp").c_str());
+ m->appendFiles((scrapQFile + toString(processIDS[i]) + ".temp"), scrapQFile);
+ remove((scrapQFile + toString(processIDS[i]) + ".temp").c_str());
+
+ m->mothurOut("Done with quality files"); m->mothurOutEndLine();
+ }
+
+ m->appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
+ remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
+
+ m->mothurOut("Done with group file"); m->mothurOutEndLine();
+
+ for (int j = 0; j < fastaNames.size(); j++) {
+ m->appendFiles((fastaNames[j] + toString(processIDS[i]) + ".temp"), fastaNames[j]);
+ remove((fastaNames[j] + toString(processIDS[i]) + ".temp").c_str());
+ }
+
+ if(qFileName != ""){
+ for (int j = 0; j < qualNames.size(); j++) {
+ m->appendFiles((qualNames[j] + toString(processIDS[i]) + ".temp"), qualNames[j]);
+ remove((qualNames[j] + toString(processIDS[i]) + ".temp").c_str());
+ }
+ }
+
+ if (allFiles) { m->mothurOut("Done with allfiles"); m->mothurOutEndLine(); }
+ }
+
return exitCommand;
#endif
}
/**************************************************************************************************/
-int TrimSeqsCommand::setLines(string filename, vector<linePair*>& lines) {
+int TrimSeqsCommand::setLines(string filename, string qfilename, vector<unsigned long int>& fastaFilePos, vector<unsigned long int>& qfileFilePos) {
try {
- lines.clear();
+ //set file positions for fasta file
+ fastaFilePos = m->divideFile(filename, processors);
- vector<unsigned long int> positions;
+ if (qfilename == "") { return processors; }
+
+ //get name of first sequence in each chunk
+ map<string, int> firstSeqNames;
+ for (int i = 0; i < (fastaFilePos.size()-1); i++) {
+ ifstream in;
+ m->openInputFile(filename, in);
+ in.seekg(fastaFilePos[i]);
+
+ Sequence temp(in);
+ firstSeqNames[temp.getName()] = i;
+
+ in.close();
+ }
+
+ //seach for filePos of each first name in the qfile and save in qfileFilePos
+ ifstream inQual;
+ m->openInputFile(qfilename, inQual);
- ifstream inFASTA;
- openInputFile(filename, inFASTA);
-
string input;
- while(!inFASTA.eof()){
- input = getline(inFASTA);
+ while(!inQual.eof()){
+ input = m->getline(inQual);
if (input.length() != 0) {
- if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
+ if(input[0] == '>'){ //this is a sequence name line
+ istringstream nameStream(input);
+
+ string sname = ""; nameStream >> sname;
+ sname = sname.substr(1);
+
+ map<string, int>::iterator it = firstSeqNames.find(sname);
+
+ if(it != firstSeqNames.end()) { //this is the start of a new chunk
+ unsigned long int pos = inQual.tellg();
+ qfileFilePos.push_back(pos - input.length() - 1);
+ firstSeqNames.erase(it);
+ }
+ }
}
+
+ if (firstSeqNames.size() == 0) { break; }
}
- inFASTA.close();
+ inQual.close();
- int numFastaSeqs = positions.size();
-
+
+ if (firstSeqNames.size() != 0) {
+ for (map<string, int>::iterator it = firstSeqNames.begin(); it != firstSeqNames.end(); it++) {
+ m->mothurOut(it->first + " is in your fasta file and not in your quality file, not using quality file."); m->mothurOutEndLine();
+ }
+ qFileName = "";
+ return processors;
+ }
+
+ //get last file position of qfile
FILE * pFile;
unsigned long int size;
//get num bytes in file
- pFile = fopen (filename.c_str(),"rb");
+ pFile = fopen (qfilename.c_str(),"rb");
if (pFile==NULL) perror ("Error opening file");
else{
fseek (pFile, 0, SEEK_END);
fclose (pFile);
}
- int numSeqsPerProcessor = numFastaSeqs / processors;
-
- for (int i = 0; i < processors; i++) {
-
- unsigned long int startPos = positions[ i * numSeqsPerProcessor ];
- if(i == processors - 1){
- numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
- }else{
- unsigned long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
- }
- lines.push_back(new linePair(startPos, numSeqsPerProcessor));
- }
+ qfileFilePos.push_back(size);
- return numFastaSeqs;
+ return processors;
}
catch(exception& e) {
m->errorOut(e, "TrimSeqsCommand", "setLines");
void TrimSeqsCommand::getOligos(vector<string>& outFASTAVec, vector<string>& outQualVec){
try {
ifstream inOligos;
- openInputFile(oligoFile, inOligos);
+ m->openInputFile(oligoFile, inOligos);
ofstream test;
//int indexPrimer = 0;
while(!inOligos.eof()){
- inOligos >> type;
+ inOligos >> type; m->gobble(inOligos);
if(type[0] == '#'){
while (!inOligos.eof()) { char c = inOligos.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
map<string, int>::iterator itPrime = primers.find(oligo);
if (itPrime != primers.end()) { m->mothurOut("primer " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
- primers[oligo]=index; index++;
- groupVector.push_back(group);
+ primers[oligo]=index; index++;
+ groupVector.push_back(group);
- if(allFiles){
- outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
- if(qFileName != ""){
- outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
- }
- if (group == "") { //if there is not a group for this primer, then this file will not get written to, but we add it to keep the indexes correct
- filesToRemove.insert((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
+ if(allFiles){
+ outFASTAVec.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
if(qFileName != ""){
- filesToRemove.insert((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
+ outQualVec.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ }
+ if (group == "") { //if there is not a group for this primer, then this file will not get written to, but we add it to keep the indexes correct
+ filesToRemove.insert((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
+ if(qFileName != ""){
+ filesToRemove.insert((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ }
+ }else {
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
+ outputTypes["fasta"].push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
+ if(qFileName != ""){
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ outputTypes["qual"].push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ }
}
- }else {
- outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
- if(qFileName != ""){
- outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
- }
}
- }
-
+
}
else if(type == "REVERSE"){
Sequence oligoRC("reverse", oligo);
map<string, int>::iterator itBar = barcodes.find(oligo);
if (itBar != barcodes.end()) { m->mothurOut("barcode " + oligo + " is in your oligos file already."); m->mothurOutEndLine(); }
- barcodes[oligo]=index; index++;
- groupVector.push_back(group);
+ barcodes[oligo]=index; index++;
+ groupVector.push_back(group);
+
+ if(allFiles){
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
+ outFASTAVec.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + group + ".fasta"));
+ if(qFileName != ""){
+ outQualVec.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ outputTypes["qual"].push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + group + ".qual"));
+ }
+ }
- if(allFiles){
- outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
- outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
- if(qFileName != ""){
- outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
- outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + group + ".qual"));
- }
- }
}else{ m->mothurOut(type + " is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring " + oligo + "."); m->mothurOutEndLine(); }
}
- gobble(inOligos);
+ m->gobble(inOligos);
}
inOligos.close();
for (map<string, int>::iterator itBar = barcodes.begin(); itBar != barcodes.end(); itBar++) {
for (map<string, int>::iterator itPrime = primers.begin(); itPrime != primers.end(); itPrime++) {
if (groupVector[itPrime->second] != "") { //there is a group for this primer
- outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
- outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
+ outputTypes["fasta"].push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
+ outFASTAVec.push_back((outputDir + m->getRootName(m->getSimpleName(fastaFile)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".fasta"));
combos[(groupVector[itBar->second] + "." + groupVector[itPrime->second])] = outFASTAVec.size()-1;
if(qFileName != ""){
- outQualVec.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
- outputNames.push_back((outputDir + getRootName(getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
+ outQualVec.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
+ outputNames.push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
+ outputTypes["qual"].push_back((outputDir + m->getRootName(m->getSimpleName(qFileName)) + groupVector[itBar->second] + "." + groupVector[itPrime->second] + ".qual"));
}
}
}
else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
- else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
-
+ else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
}
+
}
return countDiffs;