#include "chimeraslayercommand.h"
#include "chimeraslayer.h"
+#include "deconvolutecommand.h"
-
+//**********************************************************************************************************************
+vector<string> ChimeraSlayerCommand::getValidParameters(){
+ try {
+ string AlignArray[] = {"fasta", "processors","trim","split", "name","window", "include","template","numwanted", "ksize", "match","mismatch",
+ "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" };
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "getValidParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+ChimeraSlayerCommand::ChimeraSlayerCommand(){
+ try {
+ abort = true; calledHelp = true;
+ vector<string> tempOutNames;
+ outputTypes["chimera"] = tempOutNames;
+ outputTypes["accnos"] = tempOutNames;
+ outputTypes["fasta"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ChimeraSlayerCommand::getRequiredParameters(){
+ try {
+ string AlignArray[] = {"template","fasta"};
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "getRequiredParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ChimeraSlayerCommand::getRequiredFiles(){
+ try {
+ vector<string> myArray;
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "getRequiredFiles");
+ exit(1);
+ }
+}
//***************************************************************************************************************
-
ChimeraSlayerCommand::ChimeraSlayerCommand(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 {
//valid paramters for this command
- string Array[] = {"fasta", "processors", "window", "template","numwanted", "ksize", "match","mismatch",
+ string Array[] = {"fasta", "processors","name", "include","trim", "split","window", "template","numwanted", "ksize", "match","mismatch",
"divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" };
vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ vector<string> tempOutNames;
+ outputTypes["chimera"] = tempOutNames;
+ outputTypes["accnos"] = tempOutNames;
+ outputTypes["fasta"] = 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 = ""; }
- else {
- string path;
- it = parameters.find("template");
- //user has given a template file
- if(it != parameters.end()){
- path = hasPath(it->second);
- //if the user has not given a path then, add inputdir. else leave path alone.
- if (path == "") { parameters["template"] = inputDir + it->second; }
- }
- }
-
-
+
//check for required parameters
fastafile = validParameter.validFile(parameters, "fasta", false);
- if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
+ if (fastafile == "not found") { fastafile = ""; m->mothurOut("[ERROR]: fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
else {
- splitAtDash(fastafile, fastaFileNames);
+ m->splitAtDash(fastafile, fastaFileNames);
//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 = hasPath(fastaFileNames[i]);
+ 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;
- #ifdef USE_MPI
- int pid;
- MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
- MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
- if (pid == 0) {
- #endif
-
- ableToOpen = openInputFile(fastaFileNames[i], in, "noerror");
+ 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() + getSimpleName(fastaFileNames[i]);
+ string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]);
m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ableToOpen = openInputFile(tryPath, in, "noerror");
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
fastaFileNames[i] = tryPath;
}
}
- in.close();
- #ifdef USE_MPI
- for (int j = 1; j < processors; j++) {
- MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
- }
- }else{
- MPI_Status status;
- MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
+ 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;
}
-
- #endif
-
+ }
+
+ in.close();
+
if (ableToOpen == 1) {
m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
//erase from file list
}
//make sure there is at least one valid file left
- if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; }
+ if (fastaFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid files."); m->mothurOutEndLine(); abort = true; }
}
+
+ //check for required parameters
+ bool hasName = true;
+ namefile = validParameter.validFile(parameters, "name", false);
+ if (namefile == "not found") { namefile = ""; hasName = false; }
+ else {
+ m->splitAtDash(namefile, nameFileNames);
+
+ //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;
+ }
+ }
+
+ 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--;
+ }
+ }
+
+ //make sure there is at least one valid file left
+ if (nameFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid name files."); m->mothurOutEndLine(); abort = true; }
+ }
+
+ if (hasName && (nameFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of namefiles does not match the number of fastafiles, please correct."); m->mothurOutEndLine(); 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 = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
+
+
+ string path;
+ it = parameters.find("template");
+ //user has given a template file
+ if(it != parameters.end()){
+ if (it->second == "self") { templatefile = "self"; }
+ else {
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["template"] = inputDir + it->second; }
+
+ templatefile = validParameter.validFile(parameters, "template", true);
+ if (templatefile == "not open") { abort = true; }
+ else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
+ }
}
-
- templatefile = validParameter.validFile(parameters, "template", true);
- if (templatefile == "not open") { abort = true; }
- else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
-
+
string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
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);
convert(temp, parents);
temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "f"; }
- realign = isTrue(temp);
+ 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"; }
+ trimera = m->isTrue(temp);
search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "distance"; }
m->mothurOut("The chimera.slayer command reads a fastafile and templatefile and outputs potentially chimeric sequences.\n");
m->mothurOut("This command was modeled after the chimeraSlayer written by the Broad Institute.\n");
- m->mothurOut("The chimera.slayer command parameters are fasta, template, processors, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment and numwanted.\n"); //realign,
+ m->mothurOut("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,
m->mothurOut("The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required. \n");
+ m->mothurOut("The name parameter allows you to provide a name file, if you are using template=self. \n");
m->mothurOut("You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n");
- m->mothurOut("The template parameter allows you to enter a template file containing known non-chimeric sequences, and is required. \n");
+ m->mothurOut("The template parameter allows you to enter a template 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");
m->mothurOut("The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n");
#ifdef USE_MPI
m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n");
#endif
+ m->mothurOut("The trim parameter allows you to output a new fasta file containing your sequences with the chimeric ones trimmed to include only their longest piece, default=F. \n");
+ m->mothurOut("The split parameter allows you to check both pieces of non-chimeric sequence for chimeras, thus looking for trimeras and quadmeras. default=F. \n");
m->mothurOut("The window parameter allows you to specify the window size for searching for chimeras, default=50. \n");
m->mothurOut("The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default=5.\n");
m->mothurOut("The numwanted parameter allows you to specify how many sequences you would each query sequence compared with, default=15.\n");
int ChimeraSlayerCommand::execute(){
try{
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
for (int s = 0; s < fastaFileNames.size(); s++) {
int start = time(NULL);
- chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
-
- string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.chimeras";
- string accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.accnos";
+ 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 (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);
+ }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
+ string inputString = "fasta=" + fastaFileNames[s];
+ m->mothurOut("/******************************************/"); m->mothurOutEndLine();
+ m->mothurOut("Running command: unique.seqs(" + inputString + ")"); m->mothurOutEndLine();
+
+ Command* uniqueCommand = new DeconvoluteCommand(inputString);
+ uniqueCommand->execute();
+
+ map<string, vector<string> > filenames = uniqueCommand->getOutputFiles();
+
+ delete uniqueCommand;
+
+ m->mothurOut("/******************************************/"); m->mothurOutEndLine();
+
+ string 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);
+ }
+ }
+
+ if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it
+ string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimera";
+ string accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.accnos";
+ string trimFastaFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.fasta";
if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; }
templateSeqsLength = chimera->getLength();
#ifdef USE_MPI
- int pid, end, numSeqsPerProcessor;
+ int pid, numSeqsPerProcessor;
int tag = 2001;
vector<unsigned long int> MPIPos;
MPI_File inMPI;
MPI_File outMPI;
MPI_File outMPIAccnos;
+ MPI_File outMPIFasta;
int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
int inMode=MPI_MODE_RDONLY;
char outAccnosFilename[1024];
strcpy(outAccnosFilename, accnosFileName.c_str());
+
+ char outFastaFilename[1024];
+ strcpy(outFastaFilename, trimFastaFileName.c_str());
char inFileName[1024];
strcpy(inFileName, fastaFileNames[s].c_str());
MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI);
MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos);
+ if (trim) { MPI_File_open(MPI_COMM_WORLD, outFastaFilename, outMode, MPI_INFO_NULL, &outMPIFasta); }
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); 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; }
if (pid == 0) { //you are the root process
m->mothurOutEndLine();
MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status);
delete buf2;
- MPIPos = setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs
+ MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs
//send file positions to all processes
for(int i = 1; i < processors; i++) {
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
//do your part
- driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos);
+ driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); 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; }
+ 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);
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
//do your part
- driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos);
+ driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); 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
MPI_File_close(&inMPI);
MPI_File_close(&outMPI);
- MPI_File_close(&outMPIAccnos);
+ MPI_File_close(&outMPIAccnos);
+ if (trim) { MPI_File_close(&outMPIFasta); }
MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
#else
ofstream outHeader;
- string tempHeader = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader";
- openOutputFile(tempHeader, outHeader);
+ string tempHeader = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader";
+ m->openOutputFile(tempHeader, outHeader);
chimera->printHeader(outHeader);
outHeader.close();
+ vector<unsigned long int> positions = m->divideFile(fastaFileNames[s], processors);
+
+ for (int i = 0; i < (positions.size()-1); i++) {
+ lines.push_back(new linePair(positions[i], positions[(i+1)]));
+ }
+
//break up file
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(processors == 1){
- ifstream inFASTA;
- openInputFile(fastaFileNames[s], inFASTA);
- getNumSeqs(inFASTA, numSeqs);
- inFASTA.close();
-
- lines.push_back(new linePair(0, numSeqs));
-
- driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName);
-
- if (m->control_pressed) {
- remove(outputFileName.c_str());
- remove(tempHeader.c_str());
- remove(accnosFileName.c_str());
- for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); }
- for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
- delete chimera;
- return 0;
- }
+ numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
+
+ if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } remove(outputFileName.c_str()); remove(tempHeader.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
}else{
- vector<unsigned long int> positions;
processIDS.resize(0);
- ifstream inFASTA;
- openInputFile(fastaFileNames[s], inFASTA);
-
- string input;
- while(!inFASTA.eof()){
- input = getline(inFASTA);
- if (input.length() != 0) {
- if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
- }
- }
- inFASTA.close();
-
- numSeqs = positions.size();
-
- int numSeqsPerProcessor = numSeqs / processors;
-
- for (int i = 0; i < processors; i++) {
- unsigned long int startPos = positions[ i * numSeqsPerProcessor ];
- if(i == processors - 1){
- numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor;
- }
- lines.push_back(new linePair(startPos, numSeqsPerProcessor));
- }
-
- createProcesses(outputFileName, fastaFileNames[s], accnosFileName);
+ numSeqs = createProcesses(outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str());
rename((accnosFileName + toString(processIDS[0]) + ".temp").c_str(), accnosFileName.c_str());
+ if (trim) { rename((trimFastaFileName + toString(processIDS[0]) + ".temp").c_str(), trimFastaFileName.c_str()); }
//append output files
for(int i=1;i<processors;i++){
- appendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName);
+ m->appendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName);
remove((outputFileName + toString(processIDS[i]) + ".temp").c_str());
}
//append output files
for(int i=1;i<processors;i++){
- appendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName);
+ m->appendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName);
remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str());
}
- if (m->control_pressed) {
- remove(outputFileName.c_str());
- remove(accnosFileName.c_str());
- for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); }
- for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
- delete chimera;
- return 0;
+ if (trim) {
+ for(int i=1;i<processors;i++){
+ m->appendFiles((trimFastaFileName + toString(processIDS[i]) + ".temp"), trimFastaFileName);
+ remove((trimFastaFileName + toString(processIDS[i]) + ".temp").c_str());
+ }
}
-
+
+ if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } remove(outputFileName.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
}
#else
- ifstream inFASTA;
- openInputFile(fastaFileNames[s], inFASTA);
- getNumSeqs(inFASTA, numSeqs);
- inFASTA.close();
- lines.push_back(new linePair(0, numSeqs));
+ numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
- driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName);
-
- if (m->control_pressed) {
- remove(outputFileName.c_str());
- remove(tempHeader.c_str());
- remove(accnosFileName.c_str());
- for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); }
- for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
- delete chimera;
- return 0;
- }
+ if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } remove(outputFileName.c_str()); remove(tempHeader.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
#endif
- appendFiles(outputFileName, tempHeader);
+ m->appendFiles(outputFileName, tempHeader);
remove(outputFileName.c_str());
rename(tempHeader.c_str(), outputFileName.c_str());
for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
- outputNames.push_back(outputFileName);
- outputNames.push_back(accnosFileName);
+ outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName);
+ outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName);
+ if (trim) { outputNames.push_back(trimFastaFileName); outputTypes["fasta"].push_back(trimFastaFileName); }
m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine();
}
}
//**********************************************************************************************************************
-int ChimeraSlayerCommand::driver(linePair* line, string outputFName, string filename, string accnos){
+int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos, string fasta){
try {
ofstream out;
- openOutputFile(outputFName, out);
+ m->openOutputFile(outputFName, out);
ofstream out2;
- openOutputFile(accnos, out2);
+ m->openOutputFile(accnos, out2);
+
+ ofstream out3;
+ if (trim) { m->openOutputFile(fasta, out3); }
ifstream inFASTA;
- openInputFile(filename, inFASTA);
+ m->openInputFile(filename, inFASTA);
- inFASTA.seekg(line->start);
-
- for(int i=0;i<line->numSeqs;i++){
+ inFASTA.seekg(filePos->start);
+
+ bool done = false;
+ int count = 0;
+
+ while (!done) {
- if (m->control_pressed) { return 1; }
+ if (m->control_pressed) { out.close(); out2.close(); if (trim) { out3.close(); } inFASTA.close(); return 1; }
- Sequence* candidateSeq = new Sequence(inFASTA); gobble(inFASTA);
+ 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
chimera->getChimeras(candidateSeq);
if (m->control_pressed) { delete candidateSeq; return 1; }
-
- //print results
- chimera->print(out, out2);
+
+ //if you are not chimeric, then check each half
+ data_results wholeResults = chimera->getResults();
+
+ //determine if we need to split
+ bool isChimeric = false;
+
+ if (wholeResults.flag == "yes") {
+ string chimeraFlag = "no";
+ if( (wholeResults.results[0].bsa >= minBS && wholeResults.results[0].divr_qla_qrb >= divR)
+ ||
+ (wholeResults.results[0].bsb >= minBS && wholeResults.results[0].divr_qlb_qra >= divR) ) { chimeraFlag = "yes"; }
+
+
+ if (chimeraFlag == "yes") {
+ if ((wholeResults.results[0].bsa >= minBS) || (wholeResults.results[0].bsb >= minBS)) { isChimeric = true; }
+ }
+ }
+
+ if ((!isChimeric) && trimera) {
+
+ //split sequence in half by bases
+ string leftQuery, rightQuery;
+ Sequence tempSeq(candidateSeq->getName(), candidateAligned);
+ divideInHalf(tempSeq, leftQuery, rightQuery);
+
+ //run chimeraSlayer on each piece
+ Sequence* left = new Sequence(candidateSeq->getName(), leftQuery);
+ Sequence* right = new Sequence(candidateSeq->getName(), rightQuery);
+
+ //find chimeras
+ chimera->getChimeras(left);
+ data_results leftResults = chimera->getResults();
+
+ chimera->getChimeras(right);
+ data_results rightResults = chimera->getResults();
+
+ //if either piece is chimeric then report
+ Sequence* trimmed = chimera->print(out, out2, leftResults, rightResults);
+ if (trim) { trimmed->printSequence(out3); delete trimmed; }
+
+ delete left; delete right;
+
+ }else { //already chimeric
+ //print results
+ Sequence* trimmed = chimera->print(out, out2);
+ if (trim) { trimmed->printSequence(out3); delete trimmed; }
+ }
+
+
}
+ count++;
}
delete candidateSeq;
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ unsigned long int pos = inFASTA.tellg();
+ if ((pos == -1) || (pos >= filePos->end)) { break; }
+ #else
+ if (inFASTA.eof()) { break; }
+ #endif
+
//report progress
- if((i+1) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(i+1)); m->mothurOutEndLine(); }
+ if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
}
//report progress
- if((line->numSeqs) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(line->numSeqs)); m->mothurOutEndLine(); }
+ if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
out.close();
out2.close();
+ if (trim) { out3.close(); }
inFASTA.close();
- return 0;
+ return count;
}
catch(exception& e) {
m->errorOut(e, "ChimeraSlayerCommand", "driver");
}
//**********************************************************************************************************************
#ifdef USE_MPI
-int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, vector<unsigned long int>& MPIPos){
+int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, MPI_File& outFastaMPI, vector<unsigned long int>& MPIPos){
try {
MPI_Status status;
int pid;
delete buf4;
- Sequence* candidateSeq = new Sequence(iss); gobble(iss);
+ Sequence* candidateSeq = new Sequence(iss); m->gobble(iss);
+ string candidateAligned = candidateSeq->getAligned();
if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
chimera->getChimeras(candidateSeq);
if (m->control_pressed) { delete candidateSeq; return 1; }
- //cout << "about to print" << endl;
- //print results
- bool isChimeric = chimera->print(outMPI, outAccMPI);
+
+ //if you are not chimeric, then check each half
+ data_results wholeResults = chimera->getResults();
+
+ //determine if we need to split
+ bool isChimeric = false;
+
+ if (wholeResults.flag == "yes") {
+ string chimeraFlag = "no";
+ if( (wholeResults.results[0].bsa >= minBS && wholeResults.results[0].divr_qla_qrb >= divR)
+ ||
+ (wholeResults.results[0].bsb >= minBS && wholeResults.results[0].divr_qlb_qra >= divR) ) { chimeraFlag = "yes"; }
+
+
+ if (chimeraFlag == "yes") {
+ if ((wholeResults.results[0].bsa >= minBS) || (wholeResults.results[0].bsb >= minBS)) { isChimeric = true; }
+ }
+ }
+
+ if ((!isChimeric) && trimera) {
+ //split sequence in half by bases
+ string leftQuery, rightQuery;
+ Sequence tempSeq(candidateSeq->getName(), candidateAligned);
+ divideInHalf(tempSeq, leftQuery, rightQuery);
+
+ //run chimeraSlayer on each piece
+ Sequence* left = new Sequence(candidateSeq->getName(), leftQuery);
+ Sequence* right = new Sequence(candidateSeq->getName(), rightQuery);
+
+ //find chimeras
+ chimera->getChimeras(left);
+ data_results leftResults = chimera->getResults();
+
+ chimera->getChimeras(right);
+ data_results rightResults = chimera->getResults();
+
+ //if either piece is chimeric then report
+ Sequence* trimmed = chimera->print(outMPI, outAccMPI, leftResults, rightResults);
+ if (trim) {
+ string outputString = ">" + trimmed->getName() + "\n" + trimmed->getAligned() + "\n";
+ delete trimmed;
+
+ //write to accnos file
+ int length = outputString.length();
+ char* buf2 = new char[length];
+ memcpy(buf2, outputString.c_str(), length);
+
+ MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
+ delete buf2;
+ }
+
+ delete left; delete right;
+
+ }else {
+ //print results
+ Sequence* trimmed = chimera->print(outMPI, outAccMPI);
+
+ if (trim) {
+ string outputString = ">" + trimmed->getName() + "\n" + trimmed->getAligned() + "\n";
+ delete trimmed;
+
+ //write to accnos file
+ int length = outputString.length();
+ char* buf2 = new char[length];
+ memcpy(buf2, outputString.c_str(), length);
+
+ MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
+ delete buf2;
+ }
+ }
+
}
}
delete candidateSeq;
/**************************************************************************************************/
-int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos) {
+int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos, string fasta) {
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
int process = 0;
- // processIDS.resize(0);
+ int num = 0;
//loop through and create all the processes you want
while (process != processors) {
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){
- driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp");
+ num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp", fasta + toString(getpid()) + ".temp");
+
+ //pass numSeqs to parent
+ ofstream out;
+ string tempFile = outputFileName + toString(getpid()) + ".num.temp";
+ m->openOutputFile(tempFile, out);
+ out << num << endl;
+ out.close();
+
+ 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);
- }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+ }
}
//force parent to wait until all the processes are done
wait(&temp);
}
- return 0;
+ for (int i = 0; i < processIDS.size(); i++) {
+ ifstream in;
+ string tempFile = outputFileName + 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());
+ }
+
+ return num;
#endif
}
catch(exception& e) {
/**************************************************************************************************/
+int ChimeraSlayerCommand::divideInHalf(Sequence querySeq, string& leftQuery, string& rightQuery) {
+ try {
+
+ string queryUnAligned = querySeq.getUnaligned();
+ int numBases = int(queryUnAligned.length() * 0.5);
+
+ string queryAligned = querySeq.getAligned();
+ leftQuery = querySeq.getAligned();
+ rightQuery = querySeq.getAligned();
+
+ int baseCount = 0;
+ int leftSpot = 0;
+ for (int i = 0; i < queryAligned.length(); i++) {
+ //if you are a base
+ if (isalpha(queryAligned[i])) {
+ baseCount++;
+ }
+
+ //if you have half
+ if (baseCount >= numBases) { leftSpot = i; break; } //first half
+ }
+
+ //blank out right side
+ for (int i = leftSpot; i < leftQuery.length(); i++) { leftQuery[i] = '.'; }
+
+ //blank out left side
+ for (int i = 0; i < leftSpot; i++) { rightQuery[i] = '.'; }
+
+ return 0;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "divideInHalf");
+ exit(1);
+ }
+}
+
+/**************************************************************************************************/