//**********************************************************************************************************************
-
+vector<string> AlignCommand::getValidParameters(){
+ try {
+ string AlignArray[] = {"template","candidate","search","ksize","align","match","mismatch","gapopen","gapextend", "processors","flip","threshold","outputdir","inputdir"};
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "AlignCommand", "getValidParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> AlignCommand::getRequiredParameters(){
+ try {
+ string AlignArray[] = {"template","candidate"};
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "AlignCommand", "getRequiredParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> AlignCommand::getRequiredFiles(){
+ try {
+ vector<string> myArray;
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "AlignCommand", "getRequiredFiles");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+AlignCommand::AlignCommand(){
+ try {
+ abort = true; calledHelp = true;
+ vector<string> tempOutNames;
+ outputTypes["fasta"] = tempOutNames;
+ outputTypes["alignreport"] = tempOutNames;
+ outputTypes["accnos"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "AlignCommand", "AlignCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
AlignCommand::AlignCommand(string option) {
try {
-
- abort = false;
+ abort = false; calledHelp = false;
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true;}
else {
for (it = parameters.begin(); it != parameters.end(); it++) {
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
-
+
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["fasta"] = tempOutNames;
+ outputTypes["alignreport"] = tempOutNames;
+ outputTypes["accnos"] = tempOutNames;
+
//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 = ""; }
}else if (templateFileName == "not open") { abort = true; }
candidateFileName = validParameter.validFile(parameters, "candidate", false);
- if (candidateFileName == "not found") { m->mothurOut("candidate is a required parameter for the align.seqs command."); m->mothurOutEndLine(); abort = true; }
- else {
+ if (candidateFileName == "not found") {
+ //check currentFiles for a fasta file
+ if (currentFiles->getFastaFile() != "") { candidateFileName = currentFiles->getFastaFile(); m->mothurOut("Using " + candidateFileName + " as candidate file."); m->mothurOutEndLine();
+ }else { m->mothurOut("candidate is a required parameter for the align.seqs command."); m->mothurOutEndLine(); abort = true; }
+ }else {
m->splitAtDash(candidateFileName, candidateFileNames);
//go through files and make sure they are good, if not, then disregard them
int ableToOpen;
ifstream in;
-
ableToOpen = m->openInputFile(candidateFileNames[i], in, "noerror");
-
+ in.close();
+
//if you can't open it, try default location
if (ableToOpen == 1) {
if (m->getDefaultPath() != "") { //default path is set
string tryPath = m->getDefaultPath() + m->getSimpleName(candidateFileNames[i]);
m->mothurOut("Unable to open " + candidateFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ableToOpen = m->openInputFile(tryPath, in, "noerror");
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ candidateFileNames[i] = tryPath;
+ }
+ }
+
+ //if you can't open it, try output location
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(candidateFileNames[i]);
+ m->mothurOut("Unable to open " + candidateFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
candidateFileNames[i] = tryPath;
}
}
- in.close();
+
+
if (ableToOpen == 1) {
m->mothurOut("Unable to open " + candidateFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
exit(1);
}
}
-
//**********************************************************************************************************************
AlignCommand::~AlignCommand(){
int AlignCommand::execute(){
try {
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
templateDB = new AlignmentDB(templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch);
int longestBase = templateDB->getLongestBase();
m->mothurOutEndLine();
alignment = new NeedlemanOverlap(gapOpen, match, misMatch, longestBase);
}
- vector<string> outputNames;
for (int s = 0; s < candidateFileNames.size(); s++) {
- if (m->control_pressed) { return 0; }
+ if (m->control_pressed) { outputTypes.clear(); return 0; }
m->mothurOut("Aligning sequences from " + candidateFileNames[s] + " ..." ); m->mothurOutEndLine();
int start = time(NULL);
#ifdef USE_MPI
- int pid, end, numSeqsPerProcessor;
+ int pid, numSeqsPerProcessor;
int tag = 2001;
vector<unsigned long int> MPIPos;
MPIWroteAccnos = false;
MPI_File_open(MPI_COMM_WORLD, outReportFilename, outMode, MPI_INFO_NULL, &outMPIReport);
MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPIAlign); MPI_File_close(&outMPIReport); MPI_File_close(&outMPIAccnos); return 0; }
+ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPIAlign); MPI_File_close(&outMPIReport); MPI_File_close(&outMPIAccnos); outputTypes.clear(); return 0; }
if (pid == 0) { //you are the root process
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPIAlign, outMPIReport, outMPIAccnos, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPIAlign); MPI_File_close(&outMPIReport); MPI_File_close(&outMPIAccnos); return 0; }
+ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPIAlign); MPI_File_close(&outMPIReport); MPI_File_close(&outMPIAccnos); outputTypes.clear(); return 0; }
for (int i = 1; i < processors; i++) {
bool tempResult;
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPIAlign, outMPIReport, outMPIAccnos, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPIAlign); MPI_File_close(&outMPIReport); MPI_File_close(&outMPIAccnos); return 0; }
+ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPIAlign); MPI_File_close(&outMPIReport); MPI_File_close(&outMPIAccnos); outputTypes.clear(); return 0; }
MPI_Send(&MPIWroteAccnos, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
}
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(processors == 1){
numFastaSeqs = driver(lines[0], alignFileName, reportFileName, accnosFileName, candidateFileNames[s]);
-
- if (m->control_pressed) { remove(accnosFileName.c_str()); remove(alignFileName.c_str()); remove(reportFileName.c_str()); return 0; }
-
- //delete accnos file if its blank else report to user
- if (m->isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; }
- else {
- m->mothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + ".");
- if (!flip) {
- m->mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well.");
- }else{ m->mothurOut(" If the reverse compliment proved to be better it was reported."); }
- m->mothurOutEndLine();
- }
}else{
- processIDS.resize(0);
-
numFastaSeqs = createProcesses(alignFileName, reportFileName, accnosFileName, candidateFileNames[s]);
-
- rename((alignFileName + toString(processIDS[0]) + ".temp").c_str(), alignFileName.c_str());
- rename((reportFileName + toString(processIDS[0]) + ".temp").c_str(), reportFileName.c_str());
-
- //append alignment and report files
- for(int i=1;i<processors;i++){
- appendAlignFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName);
- remove((alignFileName + toString(processIDS[i]) + ".temp").c_str());
-
- appendReportFiles((reportFileName + toString(processIDS[i]) + ".temp"), reportFileName);
- remove((reportFileName + toString(processIDS[i]) + ".temp").c_str());
- }
-
- vector<string> nonBlankAccnosFiles;
- //delete blank accnos files generated with multiple processes
- for(int i=0;i<processors;i++){
- if (!(m->isBlank(accnosFileName + toString(processIDS[i]) + ".temp"))) {
- nonBlankAccnosFiles.push_back(accnosFileName + toString(processIDS[i]) + ".temp");
- }else { remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str()); }
- }
-
- //append accnos files
- if (nonBlankAccnosFiles.size() != 0) {
- rename(nonBlankAccnosFiles[0].c_str(), accnosFileName.c_str());
-
- for (int h=1; h < nonBlankAccnosFiles.size(); h++) {
- appendAlignFiles(nonBlankAccnosFiles[h], accnosFileName);
- remove(nonBlankAccnosFiles[h].c_str());
- }
- m->mothurOut("Some of you sequences generated alignments that eliminated too many bases, a list is provided in " + accnosFileName + ".");
- if (!flip) {
- m->mothurOut(" If you set the flip parameter to true mothur will try aligning the reverse compliment as well.");
- }else{ m->mothurOut(" If the reverse compliment proved to be better it was reported."); }
- m->mothurOutEndLine();
- }else{ hasAccnos = false; }
-
- if (m->control_pressed) { remove(accnosFileName.c_str()); remove(alignFileName.c_str()); remove(reportFileName.c_str()); return 0; }
}
#else
numFastaSeqs = driver(lines[0], alignFileName, reportFileName, accnosFileName, candidateFileNames[s]);
+ #endif
- if (m->control_pressed) { remove(accnosFileName.c_str()); remove(alignFileName.c_str()); remove(reportFileName.c_str()); return 0; }
+ if (m->control_pressed) { remove(accnosFileName.c_str()); remove(alignFileName.c_str()); remove(reportFileName.c_str()); outputTypes.clear(); return 0; }
//delete accnos file if its blank else report to user
if (m->isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; }
}else{ m->mothurOut(" If the reverse compliment proved to be better it was reported."); }
m->mothurOutEndLine();
}
-
- #endif
#endif
if (pid == 0) { //only one process should output to screen
#endif
- outputNames.push_back(alignFileName);
- outputNames.push_back(reportFileName);
- if (hasAccnos) { outputNames.push_back(accnosFileName); }
+ outputNames.push_back(alignFileName); outputTypes["fasta"].push_back(alignFileName);
+ outputNames.push_back(reportFileName); outputTypes["alignreport"].push_back(reportFileName);
+ if (hasAccnos) { outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName); }
#ifdef USE_MPI
}
m->mothurOutEndLine();
}
-
+ //set align file as new current fastafile
+ string currentFasta = "";
+ itTypes = outputTypes.find("fasta");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { currentFasta = (itTypes->second)[0]; }
+ }
+ currentFiles->setFastaFile(currentFasta);
+ cout << "current fasta = " << currentFiles->getFastaFile() << endl;
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
if (m->control_pressed) { return 0; }
Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
-
+ report.setCandidate(candidateSeq);
+
int origNumBases = candidateSeq->getNumBases();
string originalUnaligned = candidateSeq->getUnaligned();
int numBasesNeeded = origNumBases * threshold;
Sequence temp = templateDB->findClosestSequence(candidateSeq);
Sequence* templateSeq = &temp;
-
+
float searchScore = templateDB->getSearchScore();
Nast* nast = new Nast(alignment, candidateSeq, templateSeq);
+
Sequence* copy;
Nast* nast2;
string wasBetter = "";
//if the user wants you to try the reverse
if (flip) {
+
//get reverse compliment
copy = new Sequence(candidateSeq->getName(), originalUnaligned);
copy->reverseComplement();
accnosFile << candidateSeq->getName() << wasBetter << endl;
}
- report.setCandidate(candidateSeq);
report.setTemplate(templateSeq);
report.setSearchParameters(search, searchScore);
report.setAlignmentParameters(align, alignment);
istringstream iss (tempBuf,istringstream::in);
Sequence* candidateSeq = new Sequence(iss);
-
+ report.setCandidate(candidateSeq);
+
int origNumBases = candidateSeq->getNumBases();
string originalUnaligned = candidateSeq->getUnaligned();
int numBasesNeeded = origNumBases * threshold;
delete nast;
nast = nast2;
needToDeleteCopy = true;
+ wasBetter = "\treverse complement produced a better alignment, so mothur used the reverse complement.";
}else{
wasBetter = "\treverse complement did NOT produce a better alignment, please check sequence.";
delete nast2;
MPIWroteAccnos = true;
}
- report.setCandidate(candidateSeq);
report.setTemplate(templateSeq);
report.setSearchParameters(search, searchScore);
report.setAlignmentParameters(align, alignment);
int AlignCommand::createProcesses(string alignFileName, string reportFileName, string accnosFName, string filename) {
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- int process = 0;
+ processIDS.resize(0);
+ int process = 1;
int num = 0;
// processIDS.resize(0);
out.close();
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);
+ }
}
+ //do my part
+ num = driver(lines[0], alignFileName, reportFileName, accnosFName, filename);
+
//force parent to wait until all the processes are done
for (int i=0;i<processors;i++) {
int temp = processIDS[i];
wait(&temp);
}
+ vector<string> nonBlankAccnosFiles;
+ if (!(m->isBlank(accnosFName))) { nonBlankAccnosFiles.push_back(accnosFName); }
+ else { remove(accnosFName.c_str()); } //remove so other files can be renamed to it
+
for (int i = 0; i < processIDS.size(); i++) {
ifstream in;
string tempFile = alignFileName + toString(processIDS[i]) + ".num.temp";
m->openInputFile(tempFile, in);
if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
in.close(); remove(tempFile.c_str());
+
+ appendAlignFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName);
+ remove((alignFileName + toString(processIDS[i]) + ".temp").c_str());
+
+ appendReportFiles((reportFileName + toString(processIDS[i]) + ".temp"), reportFileName);
+ remove((reportFileName + toString(processIDS[i]) + ".temp").c_str());
+
+ if (!(m->isBlank(accnosFName + toString(processIDS[i]) + ".temp"))) {
+ nonBlankAccnosFiles.push_back(accnosFName + toString(processIDS[i]) + ".temp");
+ }else { remove((accnosFName + toString(processIDS[i]) + ".temp").c_str()); }
+
+ }
+
+ //append accnos files
+ if (nonBlankAccnosFiles.size() != 0) {
+ rename(nonBlankAccnosFiles[0].c_str(), accnosFName.c_str());
+
+ for (int h=1; h < nonBlankAccnosFiles.size(); h++) {
+ appendAlignFiles(nonBlankAccnosFiles[h], accnosFName);
+ remove(nonBlankAccnosFiles[h].c_str());
+ }
+ }else { //recreate the accnosfile if needed
+ ofstream out;
+ m->openOutputFile(accnosFName, out);
+ out.close();
}
return num;