#include "chimeraslayercommand.h"
#include "chimeraslayer.h"
+#include "deconvolutecommand.h"
+#include "referencedb.h"
+//**********************************************************************************************************************
+vector<string> ChimeraSlayerCommand::setParameters(){
+ try {
+ CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptemplate);
+ CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta);
+ CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
+ CommandParameter pwindow("window", "Number", "", "50", "", "", "",false,false); parameters.push_back(pwindow);
+ CommandParameter pksize("ksize", "Number", "", "7", "", "", "",false,false); parameters.push_back(pksize);
+ CommandParameter pmatch("match", "Number", "", "5.0", "", "", "",false,false); parameters.push_back(pmatch);
+ CommandParameter pmismatch("mismatch", "Number", "", "-4.0", "", "", "",false,false); parameters.push_back(pmismatch);
+ CommandParameter pminsim("minsim", "Number", "", "90", "", "", "",false,false); parameters.push_back(pminsim);
+ CommandParameter pmincov("mincov", "Number", "", "70", "", "", "",false,false); parameters.push_back(pmincov);
+ CommandParameter pminsnp("minsnp", "Number", "", "10", "", "", "",false,false); parameters.push_back(pminsnp);
+ CommandParameter pminbs("minbs", "Number", "", "90", "", "", "",false,false); parameters.push_back(pminbs);
+ CommandParameter psearch("search", "Multiple", "kmer-blast", "blast", "", "", "",false,false); parameters.push_back(psearch);
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
+ 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);
+ CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters);
+ CommandParameter pdivergence("divergence", "Number", "", "1.007", "", "", "",false,false); parameters.push_back(pdivergence);
+ CommandParameter pparents("parents", "Number", "", "3", "", "", "",false,false); parameters.push_back(pparents);
+ CommandParameter pincrement("increment", "Number", "", "5", "", "", "",false,false); parameters.push_back(pincrement);
+ CommandParameter pblastlocation("blastlocation", "String", "", "", "", "", "",false,false); parameters.push_back(pblastlocation);
+ CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
+ CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
+ CommandParameter psave("save", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psave);
+ vector<string> myArray;
+ for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "setParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string ChimeraSlayerCommand::getHelpString(){
+ try {
+ 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, numwanted, blastlocation 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 reference=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 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";
+#endif
+ helpString += "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";
+ helpString += "The split parameter allows you to check both pieces of non-chimeric sequence for chimeras, thus looking for trimeras and quadmeras. default=F. \n";
+ helpString += "The window parameter allows you to specify the window size for searching for chimeras, default=50. \n";
+ helpString += "The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default=5.\n";
+ helpString += "The numwanted parameter allows you to specify how many sequences you would each query sequence compared with, default=15.\n";
+ helpString += "The ksize parameter allows you to input kmersize, default is 7, used if search is kmer. \n";
+ helpString += "The match parameter allows you to reward matched bases in blast search, default is 5. \n";
+ helpString += "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";
+ helpString += "The mismatch parameter allows you to penalize mismatched bases in blast search, default is -4. \n";
+ helpString += "The divergence parameter allows you to set a cutoff for chimera determination, default is 1.007. \n";
+ helpString += "The iters parameter allows you to specify the number of bootstrap iters to do with the chimeraslayer method, default=1000.\n";
+ helpString += "The minsim parameter allows you to specify a minimum similarity with the parent fragments, default=90. \n";
+ helpString += "The mincov parameter allows you to specify minimum coverage by closest matches found in template. Default is 70, meaning 70%. \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: 10) \n";
+ helpString += "The search parameter allows you to specify search method for finding the closest parent. Choices are blast, and kmer, default blast. \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 blastlocation parameter allows you to specify the location of your blast executable. By default mothur will look in ./blast/bin relative to mothur's executable. \n";
+ helpString += "If the save parameter is set to true the reference sequences will be saved in memory, to clear them later you can use the clear.memory command. Default=f.";
+ 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";
+ helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n";
+ return helpString;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "getHelpString");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+ChimeraSlayerCommand::ChimeraSlayerCommand(){
+ try {
+ abort = true; calledHelp = true;
+ setParameters();
+ vector<string> tempOutNames;
+ outputTypes["chimera"] = tempOutNames;
+ outputTypes["accnos"] = tempOutNames;
+ outputTypes["fasta"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
+ exit(1);
+ }
+}
//***************************************************************************************************************
-
ChimeraSlayerCommand::ChimeraSlayerCommand(string option) {
try {
- abort = false;
+ abort = false; calledHelp = false;
+ ReferenceDB* rdb = ReferenceDB::getInstance();
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true; }
+ else if(option == "citation") { citation(); abort = true; calledHelp = true;}
else {
- //valid paramters for this command
- string Array[] = {"fasta", "processors", "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)));
+ vector<string> myArray = setParameters();
OptionParser parser(option);
map<string,string> parameters = parser.getParameters();
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 = 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; }
- }
- }
-
-
+
//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; }
- else {
+ if (fastafile == "not found") {
+ //if there is a current fasta file, use it
+ string filename = m->getFastaFile();
+ if (filename != "") { fastaFileNames.push_back(filename); m->mothurOut("Using " + filename + " as input file for the fasta parameter."); m->mothurOutEndLine(); }
+ else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; }
+ }else {
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 = 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]; }
+
+ 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--;
+ }
}
-
- int ableToOpen;
- ifstream in;
- ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
+ 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]; }
+ }
+
+ 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;
+ }
+ }
+
+ 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--;
+ }else {
+ m->setFastaFile(fastaFileNames[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();
- ableToOpen = m->openInputFile(tryPath, in, "noerror");
- fastaFileNames[i] = tryPath;
+ //make sure there is at least one valid file left
+ 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++) {
+
+ 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--;
}
}
- 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--;
+ 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--;
+ }else {
+ m->setNameFile(nameFileNames[i]);
+ }
}
}
//make sure there is at least one valid file left
- if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; }
+ 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 += m->hasPath(fastafile); //if user entered a file with a path then preserve it
- }
-
- 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"; }
+ outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
+
+ string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
+ m->setProcessors(temp);
convert(temp, processors);
+ temp = validParameter.validFile(parameters, "save", false); if (temp == "not found"){ temp = "f"; }
+ save = m->isTrue(temp);
+ rdb->save = save;
+ if (save) { //clear out old references
+ rdb->clearMemory();
+ }
+
+ string path;
+ it = parameters.find("reference");
+ //user has given a template file
+ if(it != parameters.end()){
+ if (it->second == "self") {
+ templatefile = "self";
+ if (save) {
+ m->mothurOut("[WARNING]: You can't save reference=self, ignoring save.");
+ m->mothurOutEndLine();
+ save = false;
+ }
+ }
+ else {
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["reference"] = inputDir + it->second; }
+
+ templatefile = validParameter.validFile(parameters, "reference", true);
+ if (templatefile == "not open") { abort = true; }
+ else if (templatefile == "not found") { //check for saved reference sequences
+ if (rdb->referenceSeqs.size() != 0) {
+ templatefile = "saved";
+ }else {
+ m->mothurOut("[ERROR]: You don't have any saved reference sequences and the reference parameter is a required.");
+ m->mothurOutEndLine();
+ abort = true;
+ }
+ }else { if (save) { rdb->setSavedReference(templatefile); } }
+ }
+ }else if (hasName) { templatefile = "self";
+ if (save) {
+ m->mothurOut("[WARNING]: You can't save reference=self, ignoring save.");
+ m->mothurOutEndLine();
+ save = false;
+ }
+ }
+ else {
+ if (rdb->referenceSeqs.size() != 0) {
+ templatefile = "saved";
+ }else {
+ m->mothurOut("[ERROR]: You don't have any saved reference sequences and the reference parameter is a required.");
+ m->mothurOutEndLine();
+ templatefile = ""; abort = true;
+ }
+ }
+
+
+
temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found") { temp = "7"; }
convert(temp, ksize);
temp = validParameter.validFile(parameters, "minbs", false); if (temp == "not found") { temp = "90"; }
convert(temp, minBS);
- temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "100"; }
+ temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "10"; }
convert(temp, minSNP);
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);
- search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "distance"; }
+ temp = validParameter.validFile(parameters, "trim", false); if (temp == "not found") { temp = "f"; }
+ trim = m->isTrue(temp);
- temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "100"; }
+ 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 = "blast"; }
+
+ temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "1000"; }
convert(temp, iters);
temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "5"; }
temp = validParameter.validFile(parameters, "numwanted", false); if (temp == "not found") { temp = "15"; }
convert(temp, numwanted);
+
+ blastlocation = validParameter.validFile(parameters, "blastlocation", false);
+ if (blastlocation == "not found") { blastlocation = ""; }
+ else {
+ //add / to name if needed
+ string lastChar = blastlocation.substr(blastlocation.length()-1);
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if (lastChar != "/") { blastlocation += "/"; }
+#else
+ if (lastChar != "\\") { blastlocation += "\\"; }
+#endif
+ blastlocation = m->getFullPathName(blastlocation);
+ string formatdbCommand = "";
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ formatdbCommand = blastlocation + "formatdb";
+#else
+ formatdbCommand = blastlocation + "formatdb.exe";
+#endif
+
+ //test to make sure formatdb exists
+ ifstream in;
+ formatdbCommand = m->getFullPathName(formatdbCommand);
+ int ableToOpen = m->openInputFile(formatdbCommand, in, "no error"); in.close();
+ if(ableToOpen == 1) { m->mothurOut("[ERROR]: " + formatdbCommand + " file does not exist. mothur requires formatdb.exe to run chimera.slayer."); m->mothurOutEndLine(); abort = true; }
+
+ string blastCommand = "";
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ blastCommand = blastlocation + "megablast";
+#else
+ blastCommand = blastlocation + "megablast.exe";
+#endif
+ //test to make sure formatdb exists
+ ifstream in2;
+ blastCommand = m->getFullPathName(blastCommand);
+ ableToOpen = m->openInputFile(blastCommand, in2, "no error"); in2.close();
+ if(ableToOpen == 1) { m->mothurOut("[ERROR]: " + blastCommand + " file does not exist. mothur requires blastall.exe to run chimera.slayer."); m->mothurOutEndLine(); abort = true; }
+ }
- if ((search != "distance") && (search != "blast") && (search != "kmer")) { m->mothurOut(search + " is not a valid search."); m->mothurOutEndLine(); abort = true; }
+ if ((search != "blast") && (search != "kmer")) { m->mothurOut(search + " is not a valid search."); m->mothurOutEndLine(); abort = true; }
+
+ if (hasName && (templatefile != "self")) { m->mothurOut("You have provided a namefile and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; }
}
}
catch(exception& e) {
exit(1);
}
}
-//**********************************************************************************************************************
-
-void ChimeraSlayerCommand::help(){
- try {
-
- 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 fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required. \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 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 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");
- m->mothurOut("The ksize parameter allows you to input kmersize, default is 7, used if search is kmer. \n");
- m->mothurOut("The match parameter allows you to reward matched bases in blast search, default is 5. \n");
- 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");
- m->mothurOut("The mismatch parameter allows you to penalize mismatched bases in blast search, default is -4. \n");
- m->mothurOut("The divergence parameter allows you to set a cutoff for chimera determination, default is 1.007. \n");
- m->mothurOut("The iters parameter allows you to specify the number of bootstrap iters to do with the chimeraslayer method, default=100.\n");
- m->mothurOut("The minsim parameter allows you to specify a minimum similarity with the parent fragments, default=90. \n");
- m->mothurOut("The mincov parameter allows you to specify minimum coverage by closest matches found in template. Default is 70, meaning 70%. \n");
- m->mothurOut("The minbs parameter allows you to specify minimum bootstrap support for calling a sequence chimeric. Default is 90, meaning 90%. \n");
- 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");
- 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");
- m->mothurOut("The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default false. \n");
- m->mothurOut("The chimera.slayer command should be in the following format: \n");
- m->mothurOut("chimera.slayer(fasta=yourFastaFile, template=yourTemplate, search=yourSearch) \n");
- m->mothurOut("Example: chimera.slayer(fasta=AD.align, template=core_set_aligned.imputed.fasta, search=kmer) \n");
- m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
- }
- catch(exception& e) {
- m->errorOut(e, "ChimeraSlayerCommand", "help");
- exit(1);
- }
-}
-
-//***************************************************************************************************************
-
-ChimeraSlayerCommand::~ChimeraSlayerCommand(){ /* do nothing */ }
-
//***************************************************************************************************************
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++) {
m->mothurOut("Checking sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
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 + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimeras";
+ 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, blastlocation);
+ }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
+ 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
+ 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();
+
+ nameFile = filenames["name"][0];
+ fastaFileNames[s] = filenames["fasta"][0];
+ }
+
+ //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++) { m->mothurRemove(outputNames[j]); } 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, blastlocation);
+ }
+
+ 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; }
+ if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
if (chimera->getUnaligned()) {
m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine();
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++) { m->mothurRemove(outputNames[j]); } delete chimera; return 0; }
if (pid == 0) { //you are the root process
m->mothurOutEndLine();
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, MPIPos);
+ driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
+
+ int numNoParents = chimera->getNumNoParents();
+ int temp;
+ for(int i = 1; i < processors; i++) {
+ MPI_Recv(&temp, 1, MPI_INT, 1, tag, MPI_COMM_WORLD, &status);
+ numNoParents += temp;
+ }
- 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 (numSeqs == numNoParents) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
+
+ 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++) { m->mothurRemove(outputNames[j]); } m->mothurRemove(outputFileName); m->mothurRemove(accnosFileName); 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, MPIPos);
+ //do your part
+ driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
+
+ int numNoParents = chimera->getNumNoParents();
+ MPI_Send(&numNoParents, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
- 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++) { m->mothurRemove(outputNames[j]); } 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
//break up file
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(processors == 1){
- numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName);
+ numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
+
+ int numNoParents = chimera->getNumNoParents();
+ if (numNoParents == numSeqs) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
- 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) { m->mothurRemove(trimFastaFileName); } m->mothurRemove(outputFileName); m->mothurRemove(tempHeader); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
}else{
processIDS.resize(0);
- numSeqs = 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++){
m->appendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName);
- remove((outputFileName + toString(processIDS[i]) + ".temp").c_str());
+ m->mothurRemove((outputFileName + toString(processIDS[i]) + ".temp"));
}
//append output files
for(int i=1;i<processors;i++){
m->appendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName);
- remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str());
+ m->mothurRemove((accnosFileName + toString(processIDS[i]) + ".temp"));
}
- 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);
+ m->mothurRemove((trimFastaFileName + toString(processIDS[i]) + ".temp"));
+ }
+ }
+
+ if (m->control_pressed) { outputTypes.clear(); if (trim) { m->mothurRemove(trimFastaFileName); } m->mothurRemove(outputFileName); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
}
#else
- numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName);
+ numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
+
+ int numNoParents = chimera->getNumNoParents();
+ if (numNoParents == numSeqs) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
+
- 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) { m->mothurRemove(trimFastaFileName); } m->mothurRemove(outputFileName); m->mothurRemove(tempHeader); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
#endif
m->appendFiles(outputFileName, tempHeader);
- remove(outputFileName.c_str());
+ m->mothurRemove(outputFileName);
rename(tempHeader.c_str(), outputFileName.c_str());
#endif
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();
}
+ //set accnos file as new current accnosfile
+ string current = "";
+ itTypes = outputTypes.find("accnos");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
+ }
+
+ if (trim) {
+ itTypes = outputTypes.find("fasta");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
+ }
+ }
+
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
}
//**********************************************************************************************************************
-int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos){
+int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos, string fasta){
try {
ofstream out;
m->openOutputFile(outputFName, out);
ofstream out2;
m->openOutputFile(accnos, out2);
+ ofstream out3;
+ if (trim) { m->openOutputFile(fasta, out3); }
+
ifstream inFASTA;
m->openInputFile(filename, inFASTA);
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); 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) {
m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
}else{
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 left; delete right;
+
+ }else { //already chimeric
+ //print results
+ Sequence trimmed = chimera->print(out, out2);
+ if (trim) { trimmed.printSequence(out3); }
+ }
+
+
}
- count++;
+ count++;
}
- delete candidateSeq;
- unsigned long int pos = inFASTA.tellg();
- if ((pos == -1) || (pos >= filePos->end)) { break; }
+ #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
+ delete candidateSeq;
//report progress
if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
}
out.close();
out2.close();
+ if (trim) { out3.close(); }
inFASTA.close();
return count;
}
//**********************************************************************************************************************
#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); 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";
+
+ //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";
+
+ //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.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
- num = 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 << num << '\t' << chimera->getNumNoParents() << endl;
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);
+ }
}
//force parent to wait until all the processes are done
wait(&temp);
}
+ int numNoParents = 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());
+ if (!in.eof()) { int tempNum = 0; int tempNumParents = 0; in >> tempNum >> tempNumParents; num += tempNum; numNoParents += tempNumParents; }
+ in.close(); m->mothurRemove(tempFile);
}
+ if (num == numNoParents) { m->mothurOut("[WARNING]: megablast returned 0 potential parents for all your sequences. This could be due to formatdb.exe not being setup properly, please check formatdb.log for errors."); m->mothurOutEndLine(); }
+
return num;
#endif
}
/**************************************************************************************************/
+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);
+ }
+}
+/**************************************************************************************************/
+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);
+ }
+}
+
+/**************************************************************************************************/
projects / mothur.git / blobdiff