2 * chimeraslayercommand.cpp
5 * Created by westcott on 3/31/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "chimeraslayercommand.h"
11 #include "chimeraslayer.h"
12 #include "deconvolutecommand.h"
14 //**********************************************************************************************************************
15 vector<string> ChimeraSlayerCommand::getValidParameters(){
17 string AlignArray[] = {"fasta", "processors","trim", "name","window", "include","template","numwanted", "ksize", "match","mismatch",
18 "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" };
19 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
23 m->errorOut(e, "ChimeraSlayerCommand", "getValidParameters");
27 //**********************************************************************************************************************
28 ChimeraSlayerCommand::ChimeraSlayerCommand(){
30 abort = true; calledHelp = true;
31 vector<string> tempOutNames;
32 outputTypes["chimera"] = tempOutNames;
33 outputTypes["accnos"] = tempOutNames;
34 outputTypes["fasta"] = tempOutNames;
37 m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
41 //**********************************************************************************************************************
42 vector<string> ChimeraSlayerCommand::getRequiredParameters(){
44 string AlignArray[] = {"template","fasta"};
45 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
49 m->errorOut(e, "ChimeraSlayerCommand", "getRequiredParameters");
53 //**********************************************************************************************************************
54 vector<string> ChimeraSlayerCommand::getRequiredFiles(){
56 vector<string> myArray;
60 m->errorOut(e, "ChimeraSlayerCommand", "getRequiredFiles");
64 //***************************************************************************************************************
65 ChimeraSlayerCommand::ChimeraSlayerCommand(string option) {
67 abort = false; calledHelp = false;
69 //allow user to run help
70 if(option == "help") { help(); abort = true; calledHelp = true; }
73 //valid paramters for this command
74 string Array[] = {"fasta", "processors","name", "include","trim", "window", "template","numwanted", "ksize", "match","mismatch",
75 "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" };
76 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
78 OptionParser parser(option);
79 map<string,string> parameters = parser.getParameters();
81 ValidParameters validParameter("chimera.slayer");
82 map<string,string>::iterator it;
84 //check to make sure all parameters are valid for command
85 for (it = parameters.begin(); it != parameters.end(); it++) {
86 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
89 vector<string> tempOutNames;
90 outputTypes["chimera"] = tempOutNames;
91 outputTypes["accnos"] = tempOutNames;
92 outputTypes["fasta"] = tempOutNames;
94 //if the user changes the input directory command factory will send this info to us in the output parameter
95 string inputDir = validParameter.validFile(parameters, "inputdir", false);
96 if (inputDir == "not found"){ inputDir = ""; }
98 //check for required parameters
99 fastafile = validParameter.validFile(parameters, "fasta", false);
100 if (fastafile == "not found") { fastafile = ""; m->mothurOut("[ERROR]: fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
102 m->splitAtDash(fastafile, fastaFileNames);
104 //go through files and make sure they are good, if not, then disregard them
105 for (int i = 0; i < fastaFileNames.size(); i++) {
106 if (inputDir != "") {
107 string path = m->hasPath(fastaFileNames[i]);
108 //if the user has not given a path then, add inputdir. else leave path alone.
109 if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
115 ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
117 //if you can't open it, try default location
118 if (ableToOpen == 1) {
119 if (m->getDefaultPath() != "") { //default path is set
120 string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]);
121 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
123 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
125 fastaFileNames[i] = tryPath;
129 if (ableToOpen == 1) {
130 if (m->getOutputDir() != "") { //default path is set
131 string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
132 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
134 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
136 fastaFileNames[i] = tryPath;
142 if (ableToOpen == 1) {
143 m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
144 //erase from file list
145 fastaFileNames.erase(fastaFileNames.begin()+i);
150 //make sure there is at least one valid file left
151 if (fastaFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid files."); m->mothurOutEndLine(); abort = true; }
155 //check for required parameters
157 namefile = validParameter.validFile(parameters, "name", false);
158 if (namefile == "not found") { namefile = ""; hasName = false; }
160 m->splitAtDash(namefile, nameFileNames);
162 //go through files and make sure they are good, if not, then disregard them
163 for (int i = 0; i < nameFileNames.size(); i++) {
164 if (inputDir != "") {
165 string path = m->hasPath(nameFileNames[i]);
166 //if the user has not given a path then, add inputdir. else leave path alone.
167 if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; }
173 ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror");
175 //if you can't open it, try default location
176 if (ableToOpen == 1) {
177 if (m->getDefaultPath() != "") { //default path is set
178 string tryPath = m->getDefaultPath() + m->getSimpleName(nameFileNames[i]);
179 m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
181 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
183 nameFileNames[i] = tryPath;
187 if (ableToOpen == 1) {
188 if (m->getOutputDir() != "") { //default path is set
189 string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]);
190 m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
192 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
194 nameFileNames[i] = tryPath;
200 if (ableToOpen == 1) {
201 m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
202 //erase from file list
203 nameFileNames.erase(nameFileNames.begin()+i);
208 //make sure there is at least one valid file left
209 if (nameFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid name files."); m->mothurOutEndLine(); abort = true; }
212 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; }
214 //if the user changes the output directory command factory will send this info to us in the output parameter
215 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
219 it = parameters.find("template");
220 //user has given a template file
221 if(it != parameters.end()){
222 if (it->second == "self") { templatefile = "self"; }
224 path = m->hasPath(it->second);
225 //if the user has not given a path then, add inputdir. else leave path alone.
226 if (path == "") { parameters["template"] = inputDir + it->second; }
228 templatefile = validParameter.validFile(parameters, "template", true);
229 if (templatefile == "not open") { abort = true; }
230 else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
234 string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
235 convert(temp, processors);
237 includeAbunds = validParameter.validFile(parameters, "include", false); if (includeAbunds == "not found") { includeAbunds = "greater"; }
238 if ((includeAbunds != "greater") && (includeAbunds != "greaterequal") && (includeAbunds != "all")) { includeAbunds = "greater"; m->mothurOut("Invalid include setting. options are greater, greaterequal or all. using greater."); m->mothurOutEndLine(); }
240 temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found") { temp = "7"; }
241 convert(temp, ksize);
243 temp = validParameter.validFile(parameters, "window", false); if (temp == "not found") { temp = "50"; }
244 convert(temp, window);
246 temp = validParameter.validFile(parameters, "match", false); if (temp == "not found") { temp = "5"; }
247 convert(temp, match);
249 temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found") { temp = "-4"; }
250 convert(temp, mismatch);
252 temp = validParameter.validFile(parameters, "divergence", false); if (temp == "not found") { temp = "1.007"; }
255 temp = validParameter.validFile(parameters, "minsim", false); if (temp == "not found") { temp = "90"; }
256 convert(temp, minSimilarity);
258 temp = validParameter.validFile(parameters, "mincov", false); if (temp == "not found") { temp = "70"; }
259 convert(temp, minCoverage);
261 temp = validParameter.validFile(parameters, "minbs", false); if (temp == "not found") { temp = "90"; }
262 convert(temp, minBS);
264 temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "100"; }
265 convert(temp, minSNP);
267 temp = validParameter.validFile(parameters, "parents", false); if (temp == "not found") { temp = "3"; }
268 convert(temp, parents);
270 temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "f"; }
271 realign = m->isTrue(temp);
273 temp = validParameter.validFile(parameters, "trim", false); if (temp == "not found") { temp = "f"; }
274 trim = m->isTrue(temp);
276 search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "distance"; }
278 temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "100"; }
279 convert(temp, iters);
281 temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "5"; }
282 convert(temp, increment);
284 temp = validParameter.validFile(parameters, "numwanted", false); if (temp == "not found") { temp = "15"; }
285 convert(temp, numwanted);
287 if ((search != "distance") && (search != "blast") && (search != "kmer")) { m->mothurOut(search + " is not a valid search."); m->mothurOutEndLine(); abort = true; }
290 catch(exception& e) {
291 m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
295 //**********************************************************************************************************************
297 void ChimeraSlayerCommand::help(){
300 m->mothurOut("The chimera.slayer command reads a fastafile and templatefile and outputs potentially chimeric sequences.\n");
301 m->mothurOut("This command was modeled after the chimeraSlayer written by the Broad Institute.\n");
302 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,
303 m->mothurOut("The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required. \n");
304 m->mothurOut("The name parameter allows you to provide a name file, if you are using template=self. \n");
305 m->mothurOut("You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n");
306 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");
307 m->mothurOut("The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n");
309 m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n");
311 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");
312 m->mothurOut("The window parameter allows you to specify the window size for searching for chimeras, default=50. \n");
313 m->mothurOut("The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default=5.\n");
314 m->mothurOut("The numwanted parameter allows you to specify how many sequences you would each query sequence compared with, default=15.\n");
315 m->mothurOut("The ksize parameter allows you to input kmersize, default is 7, used if search is kmer. \n");
316 m->mothurOut("The match parameter allows you to reward matched bases in blast search, default is 5. \n");
317 m->mothurOut("The parents parameter allows you to select the number of potential parents to investigate from the numwanted best matches after rating them, default is 3. \n");
318 m->mothurOut("The mismatch parameter allows you to penalize mismatched bases in blast search, default is -4. \n");
319 m->mothurOut("The divergence parameter allows you to set a cutoff for chimera determination, default is 1.007. \n");
320 m->mothurOut("The iters parameter allows you to specify the number of bootstrap iters to do with the chimeraslayer method, default=100.\n");
321 m->mothurOut("The minsim parameter allows you to specify a minimum similarity with the parent fragments, default=90. \n");
322 m->mothurOut("The mincov parameter allows you to specify minimum coverage by closest matches found in template. Default is 70, meaning 70%. \n");
323 m->mothurOut("The minbs parameter allows you to specify minimum bootstrap support for calling a sequence chimeric. Default is 90, meaning 90%. \n");
324 m->mothurOut("The minsnp parameter allows you to specify percent of SNPs to sample on each side of breakpoint for computing bootstrap support (default: 100) \n");
325 m->mothurOut("The search parameter allows you to specify search method for finding the closest parent. Choices are distance, blast, and kmer, default distance. \n");
326 m->mothurOut("The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default false. \n");
327 m->mothurOut("The chimera.slayer command should be in the following format: \n");
328 m->mothurOut("chimera.slayer(fasta=yourFastaFile, template=yourTemplate, search=yourSearch) \n");
329 m->mothurOut("Example: chimera.slayer(fasta=AD.align, template=core_set_aligned.imputed.fasta, search=kmer) \n");
330 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
332 catch(exception& e) {
333 m->errorOut(e, "ChimeraSlayerCommand", "help");
338 //***************************************************************************************************************
340 ChimeraSlayerCommand::~ChimeraSlayerCommand(){ /* do nothing */ }
342 //***************************************************************************************************************
344 int ChimeraSlayerCommand::execute(){
347 if (abort == true) { if (calledHelp) { return 0; } return 2; }
349 for (int s = 0; s < fastaFileNames.size(); s++) {
351 m->mothurOut("Checking sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
353 int start = time(NULL);
355 if (templatefile != "self") { //you want to run slayer with a refernce template
356 chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
358 if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
359 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);
362 m->mothurOutEndLine(); m->mothurOut("No namesfile given, running unique.seqs command to generate one."); m->mothurOutEndLine(); m->mothurOutEndLine();
364 //use unique.seqs to create new name and fastafile
365 string inputString = "fasta=" + fastaFileNames[s];
366 m->mothurOut("/******************************************/"); m->mothurOutEndLine();
367 m->mothurOut("Running command: unique.seqs(" + inputString + ")"); m->mothurOutEndLine();
369 Command* uniqueCommand = new DeconvoluteCommand(inputString);
370 uniqueCommand->execute();
372 map<string, vector<string> > filenames = uniqueCommand->getOutputFiles();
374 delete uniqueCommand;
376 m->mothurOut("/******************************************/"); m->mothurOutEndLine();
378 string nameFile = filenames["name"][0];
379 fastaFileNames[s] = filenames["fasta"][0];
381 chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFile, search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
385 if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it
386 string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimera";
387 string accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.accnos";
388 string trimFastaFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.fasta";
390 if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; }
392 if (chimera->getUnaligned()) {
393 m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine();
397 templateSeqsLength = chimera->getLength();
400 int pid, numSeqsPerProcessor;
402 vector<unsigned long int> MPIPos;
405 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
406 MPI_Comm_size(MPI_COMM_WORLD, &processors);
410 MPI_File outMPIAccnos;
411 MPI_File outMPIFasta;
413 int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
414 int inMode=MPI_MODE_RDONLY;
416 char outFilename[1024];
417 strcpy(outFilename, outputFileName.c_str());
419 char outAccnosFilename[1024];
420 strcpy(outAccnosFilename, accnosFileName.c_str());
422 char outFastaFilename[1024];
423 strcpy(outFastaFilename, trimFastaFileName.c_str());
425 char inFileName[1024];
426 strcpy(inFileName, fastaFileNames[s].c_str());
428 MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
429 MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI);
430 MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos);
431 if (trim) { MPI_File_open(MPI_COMM_WORLD, outFastaFilename, outMode, MPI_INFO_NULL, &outMPIFasta); }
433 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; }
435 if (pid == 0) { //you are the root process
436 m->mothurOutEndLine();
437 m->mothurOut("Only reporting sequence supported by " + toString(minBS) + "% of bootstrapped results.");
438 m->mothurOutEndLine();
440 string outTemp = "Name\tLeftParent\tRightParent\tDivQLAQRB\tPerIDQLAQRB\tBootStrapA\tDivQLBQRA\tPerIDQLBQRA\tBootStrapB\tFlag\tLeftWindow\tRightWindow\n";
443 int length = outTemp.length();
444 char* buf2 = new char[length];
445 memcpy(buf2, outTemp.c_str(), length);
447 MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status);
450 MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs
452 //send file positions to all processes
453 for(int i = 1; i < processors; i++) {
454 MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
455 MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
458 //figure out how many sequences you have to align
459 numSeqsPerProcessor = numSeqs / processors;
460 int startIndex = pid * numSeqsPerProcessor;
461 if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
464 driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
466 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; }
468 }else{ //you are a child process
469 MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
470 MPIPos.resize(numSeqs+1);
471 MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
473 //figure out how many sequences you have to align
474 numSeqsPerProcessor = numSeqs / processors;
475 int startIndex = pid * numSeqsPerProcessor;
476 if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
479 driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
481 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; }
485 MPI_File_close(&inMPI);
486 MPI_File_close(&outMPI);
487 MPI_File_close(&outMPIAccnos);
488 if (trim) { MPI_File_close(&outMPIFasta); }
489 MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
493 string tempHeader = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader";
494 m->openOutputFile(tempHeader, outHeader);
496 chimera->printHeader(outHeader);
499 vector<unsigned long int> positions = m->divideFile(fastaFileNames[s], processors);
501 for (int i = 0; i < (positions.size()-1); i++) {
502 lines.push_back(new linePair(positions[i], positions[(i+1)]));
506 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
508 numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
510 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; }
513 processIDS.resize(0);
515 numSeqs = createProcesses(outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
517 rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str());
518 rename((accnosFileName + toString(processIDS[0]) + ".temp").c_str(), accnosFileName.c_str());
519 if (trim) { rename((trimFastaFileName + toString(processIDS[0]) + ".temp").c_str(), trimFastaFileName.c_str()); }
521 //append output files
522 for(int i=1;i<processors;i++){
523 m->appendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName);
524 remove((outputFileName + toString(processIDS[i]) + ".temp").c_str());
527 //append output files
528 for(int i=1;i<processors;i++){
529 m->appendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName);
530 remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str());
534 for(int i=1;i<processors;i++){
535 m->appendFiles((trimFastaFileName + toString(processIDS[i]) + ".temp"), trimFastaFileName);
536 remove((trimFastaFileName + toString(processIDS[i]) + ".temp").c_str());
540 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; }
544 numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
546 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; }
550 m->appendFiles(outputFileName, tempHeader);
552 remove(outputFileName.c_str());
553 rename(tempHeader.c_str(), outputFileName.c_str());
559 for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
561 outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName);
562 outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName);
563 if (trim) { outputNames.push_back(trimFastaFileName); outputTypes["fasta"].push_back(trimFastaFileName); }
565 m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine();
568 m->mothurOutEndLine();
569 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
570 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
571 m->mothurOutEndLine();
576 catch(exception& e) {
577 m->errorOut(e, "ChimeraSlayerCommand", "execute");
581 //**********************************************************************************************************************
583 int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos, string fasta){
586 m->openOutputFile(outputFName, out);
589 m->openOutputFile(accnos, out2);
592 if (trim) { m->openOutputFile(fasta, out3); }
595 m->openInputFile(filename, inFASTA);
597 inFASTA.seekg(filePos->start);
604 if (m->control_pressed) { out.close(); out2.close(); if (trim) { out3.close(); } inFASTA.close(); return 1; }
606 Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
608 if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
610 if (candidateSeq->getAligned().length() != templateSeqsLength) {
611 m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
614 chimera->getChimeras(candidateSeq);
616 if (m->control_pressed) { delete candidateSeq; return 1; }
619 Sequence* trimmed = chimera->print(out, out2);
621 if (trim) { trimmed->printSequence(out3); delete trimmed; }
627 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
628 unsigned long int pos = inFASTA.tellg();
629 if ((pos == -1) || (pos >= filePos->end)) { break; }
631 if (inFASTA.eof()) { break; }
635 if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
638 if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
642 if (trim) { out3.close(); }
647 catch(exception& e) {
648 m->errorOut(e, "ChimeraSlayerCommand", "driver");
652 //**********************************************************************************************************************
654 int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, MPI_File& outFastaMPI, vector<unsigned long int>& MPIPos){
658 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
660 for(int i=0;i<num;i++){
662 if (m->control_pressed) { return 1; }
665 int length = MPIPos[start+i+1] - MPIPos[start+i];
667 char* buf4 = new char[length];
668 MPI_File_read_at(inMPI, MPIPos[start+i], buf4, length, MPI_CHAR, &status);
670 string tempBuf = buf4;
671 if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
672 istringstream iss (tempBuf,istringstream::in);
676 Sequence* candidateSeq = new Sequence(iss); m->gobble(iss);
678 if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
680 if (candidateSeq->getAligned().length() != templateSeqsLength) {
681 m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
685 chimera->getChimeras(candidateSeq);
687 if (m->control_pressed) { delete candidateSeq; return 1; }
690 Sequence* trimmed = chimera->print(outMPI, outAccMPI);
693 string outputString = ">" + trimmed->getName() + "\n" + trimmed->getAligned() + "\n";
696 //write to accnos file
697 int length = outputString.length();
698 char* buf2 = new char[length];
699 memcpy(buf2, outputString.c_str(), length);
701 MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
710 if((i+1) % 100 == 0){ cout << "Processing sequence: " << (i+1) << endl; m->mothurOutJustToLog("Processing sequence: " + toString(i+1) + "\n"); }
713 if(num % 100 != 0){ cout << "Processing sequence: " << num << endl; m->mothurOutJustToLog("Processing sequence: " + toString(num) + "\n"); }
718 catch(exception& e) {
719 m->errorOut(e, "ChimeraSlayerCommand", "driverMPI");
725 /**************************************************************************************************/
727 int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos, string fasta) {
729 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
733 //loop through and create all the processes you want
734 while (process != processors) {
738 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
741 num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp", fasta + toString(getpid()) + ".temp");
743 //pass numSeqs to parent
745 string tempFile = outputFileName + toString(getpid()) + ".num.temp";
746 m->openOutputFile(tempFile, out);
752 m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
753 for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
758 //force parent to wait until all the processes are done
759 for (int i=0;i<processors;i++) {
760 int temp = processIDS[i];
764 for (int i = 0; i < processIDS.size(); i++) {
766 string tempFile = outputFileName + toString(processIDS[i]) + ".num.temp";
767 m->openInputFile(tempFile, in);
768 if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
769 in.close(); remove(tempFile.c_str());
775 catch(exception& e) {
776 m->errorOut(e, "ChimeraSlayerCommand", "createProcesses");
781 /**************************************************************************************************/