CommandParameter pminsnp("minsnp", "Number", "", "100", "", "", "",false,false); parameters.push_back(pminsnp);
CommandParameter pminbs("minbs", "Number", "", "90", "", "", "",false,false); parameters.push_back(pminbs);
CommandParameter psearch("search", "Multiple", "kmer-blast-distance", "distance", "", "", "",false,false); parameters.push_back(psearch);
- CommandParameter pinclude("include", "Multiple", "greater-greaterequal-all", "greater", "", "", "",false,false); parameters.push_back(pinclude);
CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
- CommandParameter prealign("realign", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(prealign);
+ CommandParameter prealign("realign", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(prealign);
CommandParameter ptrim("trim", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(ptrim);
CommandParameter psplit("split", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psplit);
CommandParameter pnumwanted("numwanted", "Number", "", "15", "", "", "",false,false); parameters.push_back(pnumwanted);
string helpString = "";
helpString += "The chimera.slayer command reads a fastafile and referencefile and outputs potentially chimeric sequences.\n";
helpString += "This command was modeled after the chimeraSlayer written by the Broad Institute.\n";
- helpString += "The chimera.slayer command parameters are fasta, name, template, processors, trim, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment and numwanted.\n"; //realign,
+ helpString += "The chimera.slayer command parameters are fasta, name, template, processors, trim, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment, numwanted, and realign.\n";
helpString += "The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required, unless you have a valid current fasta file. \n";
helpString += "The name parameter allows you to provide a name file, if you are using template=self. \n";
helpString += "You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n";
helpString += "The reference parameter allows you to enter a reference file containing known non-chimeric sequences, and is required. You may also set template=self, in this case the abundant sequences will be used as potential parents. \n";
- helpString += "The include parameter is used when template=self and allows you to choose which sequences will make up the \"template\". Options are greater, greaterequal and all, default=greater, meaning sequences with greater abundance than the query sequence. \n";
helpString += "The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n";
#ifdef USE_MPI
helpString += "When using MPI, the processors parameter is set to the number of MPI processes running. \n";
helpString += "The minbs parameter allows you to specify minimum bootstrap support for calling a sequence chimeric. Default is 90, meaning 90%. \n";
helpString += "The minsnp parameter allows you to specify percent of SNPs to sample on each side of breakpoint for computing bootstrap support (default: 100) \n";
helpString += "The search parameter allows you to specify search method for finding the closest parent. Choices are distance, blast, and kmer, default distance. \n";
- helpString += "The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default false. \n";
+ helpString += "The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default true. \n";
helpString += "The chimera.slayer command should be in the following format: \n";
helpString += "chimera.slayer(fasta=yourFastaFile, reference=yourTemplate, search=yourSearch) \n";
helpString += "Example: chimera.slayer(fasta=AD.align, reference=core_set_aligned.imputed.fasta, search=kmer) \n";
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < fastaFileNames.size(); i++) {
- if (inputDir != "") {
- string path = m->hasPath(fastaFileNames[i]);
- //if the user has not given a path then, add inputdir. else leave path alone.
- if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
- }
-
- int ableToOpen;
- ifstream in;
- ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
-
- //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(fastaFileNames[i]);
- m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ifstream in2;
- ableToOpen = m->openInputFile(tryPath, in2, "noerror");
- in2.close();
- fastaFileNames[i] = tryPath;
+ bool ignore = false;
+ if (fastaFileNames[i] == "current") {
+ fastaFileNames[i] = m->getFastaFile();
+ if (fastaFileNames[i] != "") { m->mothurOut("Using " + fastaFileNames[i] + " as input file for the fasta parameter where you had given current."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You have no current fastafile, ignoring current."); m->mothurOutEndLine(); ignore=true;
+ //erase from file list
+ fastaFileNames.erase(fastaFileNames.begin()+i);
+ i--;
}
}
- if (ableToOpen == 1) {
- if (m->getOutputDir() != "") { //default path is set
- string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
- m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
- ifstream in2;
- ableToOpen = m->openInputFile(tryPath, in2, "noerror");
- in2.close();
- fastaFileNames[i] = tryPath;
+ if (!ignore) {
+
+ if (inputDir != "") {
+ string path = m->hasPath(fastaFileNames[i]);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
}
- }
-
- in.close();
+
+ int ableToOpen;
+ ifstream in;
+
+ ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
- if (ableToOpen == 1) {
- m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
- //erase from file list
- fastaFileNames.erase(fastaFileNames.begin()+i);
- i--;
+ //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(fastaFileNames[i]);
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ fastaFileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ fastaFileNames[i] = tryPath;
+ }
+ }
+
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
+ //erase from file list
+ fastaFileNames.erase(fastaFileNames.begin()+i);
+ i--;
+ }
}
}
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < nameFileNames.size(); i++) {
- if (inputDir != "") {
- string path = m->hasPath(nameFileNames[i]);
- //if the user has not given a path then, add inputdir. else leave path alone.
- if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; }
- }
- int ableToOpen;
- ifstream in;
-
- ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror");
-
- //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(nameFileNames[i]);
- m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ifstream in2;
- ableToOpen = m->openInputFile(tryPath, in2, "noerror");
- in2.close();
- nameFileNames[i] = tryPath;
+ bool ignore = false;
+ if (nameFileNames[i] == "current") {
+ nameFileNames[i] = m->getNameFile();
+ if (nameFileNames[i] != "") { m->mothurOut("Using " + nameFileNames[i] + " as input file for the name parameter where you had given current."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You have no current namefile, ignoring current."); m->mothurOutEndLine(); ignore=true;
+ //erase from file list
+ nameFileNames.erase(nameFileNames.begin()+i);
+ i--;
}
}
- if (ableToOpen == 1) {
- if (m->getOutputDir() != "") { //default path is set
- string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]);
- m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
- ifstream in2;
- ableToOpen = m->openInputFile(tryPath, in2, "noerror");
- in2.close();
- nameFileNames[i] = tryPath;
+ if (!ignore) {
+
+ if (inputDir != "") {
+ string path = m->hasPath(nameFileNames[i]);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; }
+ }
+
+ int ableToOpen;
+ ifstream in;
+
+ ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror");
+
+ //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(nameFileNames[i]);
+ m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ nameFileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]);
+ m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ nameFileNames[i] = tryPath;
+ }
+ }
+
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
+ //erase from file list
+ nameFileNames.erase(nameFileNames.begin()+i);
+ i--;
}
- }
-
- in.close();
-
- if (ableToOpen == 1) {
- m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
- //erase from file list
- nameFileNames.erase(nameFileNames.begin()+i);
- i--;
}
}
m->setProcessors(temp);
convert(temp, processors);
- includeAbunds = validParameter.validFile(parameters, "include", false); if (includeAbunds == "not found") { includeAbunds = "greater"; }
- if ((includeAbunds != "greater") && (includeAbunds != "greaterequal") && (includeAbunds != "all")) { includeAbunds = "greater"; m->mothurOut("Invalid include setting. options are greater, greaterequal or all. using greater."); m->mothurOutEndLine(); }
-
temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found") { temp = "7"; }
convert(temp, ksize);
temp = validParameter.validFile(parameters, "parents", false); if (temp == "not found") { temp = "3"; }
convert(temp, parents);
- temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "f"; }
+ temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "t"; }
realign = m->isTrue(temp);
temp = validParameter.validFile(parameters, "trim", false); if (temp == "not found") { temp = "f"; }
trim = m->isTrue(temp);
- temp = validParameter.validFile(parameters, "split", false); if (temp == "not found") { temp = "f"; }
+ temp = validParameter.validFile(parameters, "split", false); if (temp == "not found") { temp = "f"; }
trimera = m->isTrue(temp);
search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "distance"; }
int ChimeraSlayerCommand::execute(){
try{
-
if (abort == true) { if (calledHelp) { return 0; } return 2; }
for (int s = 0; s < fastaFileNames.size(); s++) {
if (templatefile != "self") { //you want to run slayer with a refernce template
chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
}else {
+ if (processors != 1) { m->mothurOut("When using template=self, mothur can only use 1 processor, continuing."); m->mothurOutEndLine(); processors = 1; }
+ string nameFile = "";
if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
- chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFileNames[s], search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
+ nameFile = nameFileNames[s];
}else {
-
m->mothurOutEndLine(); m->mothurOut("No namesfile given, running unique.seqs command to generate one."); m->mothurOutEndLine(); m->mothurOutEndLine();
//use unique.seqs to create new name and fastafile
m->mothurOut("/******************************************/"); m->mothurOutEndLine();
- string nameFile = filenames["name"][0];
+ nameFile = filenames["name"][0];
fastaFileNames[s] = filenames["fasta"][0];
-
- chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFile, search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
}
+
+ //sort fastafile by abundance, returns new sorted fastafile name
+ m->mothurOut("Sorting fastafile according to abundance..."); cout.flush();
+ map<string, int> priority = sortFastaFile(fastaFileNames[s], nameFile);
+ m->mothurOut("Done."); m->mothurOutEndLine();
+
+ if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; }
+
+ chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, priority, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
}
if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it
MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs
- //send file positions to all processes
- for(int i = 1; i < processors; i++) {
- MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
- MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
+ if (templatefile != "self") { //if template=self we can only use 1 processor
+ //send file positions to all processes
+ for(int i = 1; i < processors; i++) {
+ MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
+ MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
+ }
}
-
//figure out how many sequences you have to align
numSeqsPerProcessor = numSeqs / processors;
int startIndex = pid * numSeqsPerProcessor;
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
-
+
+ if (templatefile == "self") { //if template=self we can only use 1 processor
+ startIndex = 0;
+ numSeqsPerProcessor = numSeqs;
+ }
+
//do your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } remove(outputFileName.c_str()); remove(accnosFileName.c_str()); delete chimera; return 0; }
}else{ //you are a child process
- MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
- MPIPos.resize(numSeqs+1);
- MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
+ if (templatefile != "self") { //if template=self we can only use 1 processor
+ MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
+ MPIPos.resize(numSeqs+1);
+ MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
- //figure out how many sequences you have to align
- numSeqsPerProcessor = numSeqs / processors;
- int startIndex = pid * numSeqsPerProcessor;
- if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
+ //figure out how many sequences you have to align
+ numSeqsPerProcessor = numSeqs / processors;
+ int startIndex = pid * numSeqsPerProcessor;
+ if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
- //do your part
- driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
+ //do your part
+ driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
- if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
+
+ }
}
//close files
Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
string candidateAligned = candidateSeq->getAligned();
-
+
if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
if (candidateSeq->getAligned().length() != templateSeqsLength) {
}
count++;
}
- delete candidateSeq;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
unsigned long int pos = inFASTA.tellg();
+ //cout << candidateSeq->getName() << '\t' << pos << endl;
if ((pos == -1) || (pos >= filePos->end)) { break; }
#else
if (inFASTA.eof()) { break; }
#endif
+ delete candidateSeq;
//report progress
if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
}
exit(1);
}
}
+/**************************************************************************************************/
+map<string, int> ChimeraSlayerCommand::sortFastaFile(string fastaFile, string nameFile) {
+ try {
+ map<string, int> nameAbund;
+
+ //read through fastafile and store info
+ map<string, string> seqs;
+ ifstream in;
+ m->openInputFile(fastaFile, in);
+
+ while (!in.eof()) {
+
+ if (m->control_pressed) { in.close(); return nameAbund; }
+
+ Sequence seq(in); m->gobble(in);
+ seqs[seq.getName()] = seq.getAligned();
+ }
+
+ in.close();
+
+ //read namefile
+ vector<seqPriorityNode> nameMapCount;
+ int error = m->readNames(nameFile, nameMapCount, seqs);
+
+ if (m->control_pressed) { return nameAbund; }
+
+ if (error == 1) { m->control_pressed = true; return nameAbund; }
+ if (seqs.size() != nameMapCount.size()) { m->mothurOut( "The number of sequences in your fastafile does not match the number of sequences in your namefile, aborting."); m->mothurOutEndLine(); m->control_pressed = true; return nameAbund; }
+
+ sort(nameMapCount.begin(), nameMapCount.end(), compareSeqPriorityNodes);
+
+ string newFasta = fastaFile + ".temp";
+ ofstream out;
+ m->openOutputFile(newFasta, out);
+
+ //print new file in order of
+ for (int i = 0; i < nameMapCount.size(); i++) {
+ out << ">" << nameMapCount[i].name << endl << nameMapCount[i].seq << endl;
+ nameAbund[nameMapCount[i].name] = nameMapCount[i].numIdentical;
+ }
+ out.close();
+
+ rename(newFasta.c_str(), fastaFile.c_str());
+
+ return nameAbund;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "sortFastaFile");
+ exit(1);
+ }
+}
/**************************************************************************************************/