//#include "uc.h"
#include "sequence.hpp"
#include "referencedb.h"
-
+#include "systemcommand.h"
//**********************************************************************************************************************
vector<string> ChimeraUchimeCommand::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 pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup);
- CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
- CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
- CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
- CommandParameter pabskew("abskew", "Number", "", "1.9", "", "", "",false,false); parameters.push_back(pabskew);
- CommandParameter pchimealns("chimealns", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pchimealns);
- CommandParameter pminh("minh", "Number", "", "0.3", "", "", "",false,false); parameters.push_back(pminh);
- CommandParameter pmindiv("mindiv", "Number", "", "0.5", "", "", "",false,false); parameters.push_back(pmindiv);
- CommandParameter pxn("xn", "Number", "", "8.0", "", "", "",false,false); parameters.push_back(pxn);
- CommandParameter pdn("dn", "Number", "", "1.4", "", "", "",false,false); parameters.push_back(pdn);
- CommandParameter pxa("xa", "Number", "", "1", "", "", "",false,false); parameters.push_back(pxa);
- CommandParameter pchunks("chunks", "Number", "", "4", "", "", "",false,false); parameters.push_back(pchunks);
- CommandParameter pminchunk("minchunk", "Number", "", "64", "", "", "",false,false); parameters.push_back(pminchunk);
- CommandParameter pidsmoothwindow("idsmoothwindow", "Number", "", "32", "", "", "",false,false); parameters.push_back(pidsmoothwindow);
+ CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none","",false,true,true); parameters.push_back(ptemplate);
+ CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none","chimera-accnos",false,true,true); parameters.push_back(pfasta);
+ CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "none","",false,false,true); parameters.push_back(pname);
+ CommandParameter pcount("count", "InputTypes", "", "", "NameCount-CountGroup", "none", "none","",false,false,true); parameters.push_back(pcount);
+ CommandParameter pgroup("group", "InputTypes", "", "", "CountGroup", "none", "none","",false,false,true); parameters.push_back(pgroup);
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors);
+ CommandParameter pstrand("strand", "String", "", "", "", "", "","",false,false); parameters.push_back(pstrand);
+ CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir);
+ CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir);
+ CommandParameter pabskew("abskew", "Number", "", "1.9", "", "", "","",false,false); parameters.push_back(pabskew);
+ CommandParameter pchimealns("chimealns", "Boolean", "", "F", "", "", "","alns",false,false); parameters.push_back(pchimealns);
+ CommandParameter pminh("minh", "Number", "", "0.3", "", "", "","",false,false); parameters.push_back(pminh);
+ CommandParameter pmindiv("mindiv", "Number", "", "0.5", "", "", "","",false,false); parameters.push_back(pmindiv);
+ CommandParameter pxn("xn", "Number", "", "8.0", "", "", "","",false,false); parameters.push_back(pxn);
+ CommandParameter pdn("dn", "Number", "", "1.4", "", "", "","",false,false); parameters.push_back(pdn);
+ CommandParameter pxa("xa", "Number", "", "1", "", "", "","",false,false); parameters.push_back(pxa);
+ CommandParameter pchunks("chunks", "Number", "", "4", "", "", "","",false,false); parameters.push_back(pchunks);
+ CommandParameter pminchunk("minchunk", "Number", "", "64", "", "", "","",false,false); parameters.push_back(pminchunk);
+ CommandParameter pidsmoothwindow("idsmoothwindow", "Number", "", "32", "", "", "","",false,false); parameters.push_back(pidsmoothwindow);
+ CommandParameter pdups("dereplicate", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(pdups);
+
//CommandParameter pminsmoothid("minsmoothid", "Number", "", "0.95", "", "", "",false,false); parameters.push_back(pminsmoothid);
- CommandParameter pmaxp("maxp", "Number", "", "2", "", "", "",false,false); parameters.push_back(pmaxp);
- CommandParameter pskipgaps("skipgaps", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pskipgaps);
- CommandParameter pskipgaps2("skipgaps2", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pskipgaps2);
- CommandParameter pminlen("minlen", "Number", "", "10", "", "", "",false,false); parameters.push_back(pminlen);
- CommandParameter pmaxlen("maxlen", "Number", "", "10000", "", "", "",false,false); parameters.push_back(pmaxlen);
- CommandParameter pucl("ucl", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pucl);
- CommandParameter pqueryfract("queryfract", "Number", "", "0.5", "", "", "",false,false); parameters.push_back(pqueryfract);
+ CommandParameter pmaxp("maxp", "Number", "", "2", "", "", "","",false,false); parameters.push_back(pmaxp);
+ CommandParameter pskipgaps("skipgaps", "Boolean", "", "T", "", "", "","",false,false); parameters.push_back(pskipgaps);
+ CommandParameter pskipgaps2("skipgaps2", "Boolean", "", "T", "", "", "","",false,false); parameters.push_back(pskipgaps2);
+ CommandParameter pminlen("minlen", "Number", "", "10", "", "", "","",false,false); parameters.push_back(pminlen);
+ CommandParameter pmaxlen("maxlen", "Number", "", "10000", "", "", "","",false,false); parameters.push_back(pmaxlen);
+ CommandParameter pucl("ucl", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(pucl);
+ CommandParameter pqueryfract("queryfract", "Number", "", "0.5", "", "", "","",false,false); parameters.push_back(pqueryfract);
vector<string> myArray;
for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
string helpString = "";
helpString += "The chimera.uchime command reads a fastafile and referencefile and outputs potentially chimeric sequences.\n";
helpString += "This command is a wrapper for uchime written by Robert C. Edgar.\n";
- helpString += "The chimera.uchime command parameters are fasta, name, reference, processors, abskew, chimealns, minh, mindiv, xn, dn, xa, chunks, minchunk, idsmoothwindow, minsmoothid, maxp, skipgaps, skipgaps2, minlen, maxlen, ucl and queryfact.\n";
+ helpString += "The chimera.uchime command parameters are fasta, name, count, reference, processors, dereplicate, abskew, chimealns, minh, mindiv, xn, dn, xa, chunks, minchunk, idsmoothwindow, minsmoothid, maxp, skipgaps, skipgaps2, minlen, maxlen, ucl, strand and queryfact.\n";
helpString += "The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required, unless you have a valid current fasta file. \n";
helpString += "The name parameter allows you to provide a name file, if you are using template=self. \n";
+ helpString += "The count parameter allows you to provide a count file, if you are using template=self. When you use a count file with group info and dereplicate=T, mothur will create a *.pick.count_table file containing seqeunces after chimeras are removed. \n";
helpString += "You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n";
helpString += "The group parameter allows you to provide a group file. The group file can be used with a namesfile and reference=self. When checking sequences, only sequences from the same group as the query sequence will be used as the reference. \n";
+ helpString += "If the dereplicate parameter is false, then if one group finds the seqeunce to be chimeric, then all groups find it to be chimeric, default=f.\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";
helpString += "The abskew parameter can only be used with template=self. Minimum abundance skew. Default 1.9. Abundance skew is: min [ abund(parent1), abund(parent2) ] / abund(query).\n";
}
}
//**********************************************************************************************************************
+string ChimeraUchimeCommand::getOutputPattern(string type) {
+ try {
+ string pattern = "";
+
+ if (type == "chimera") { pattern = "[filename],uchime.chimeras"; }
+ else if (type == "accnos") { pattern = "[filename],uchime.accnos"; }
+ else if (type == "alns") { pattern = "[filename],uchime.alns"; }
+ else if (type == "count") { pattern = "[filename],uchime.pick.count_table"; }
+ else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; }
+
+ return pattern;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraUchimeCommand", "getOutputPattern");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
ChimeraUchimeCommand::ChimeraUchimeCommand(){
try {
abort = true; calledHelp = true;
outputTypes["chimera"] = tempOutNames;
outputTypes["accnos"] = tempOutNames;
outputTypes["alns"] = tempOutNames;
+ outputTypes["count"] = tempOutNames;
}
catch(exception& e) {
m->errorOut(e, "ChimeraUchimeCommand", "ChimeraUchimeCommand");
//***************************************************************************************************************
ChimeraUchimeCommand::ChimeraUchimeCommand(string option) {
try {
- abort = false; calledHelp = false;
+ abort = false; calledHelp = false; hasName=false; hasCount=false;
ReferenceDB* rdb = ReferenceDB::getInstance();
//allow user to run help
outputTypes["chimera"] = tempOutNames;
outputTypes["accnos"] = tempOutNames;
outputTypes["alns"] = tempOutNames;
+ outputTypes["count"] = 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);
//check for required parameters
- bool hasName = true;
namefile = validParameter.validFile(parameters, "name", false);
- if (namefile == "not found") { namefile = ""; hasName = false; }
+ if (namefile == "not found") { namefile = ""; }
else {
m->splitAtDash(namefile, nameFileNames);
}
}
}
+ }
+
+ if (nameFileNames.size() != 0) { hasName = true; }
+
+ //check for required parameters
+ vector<string> countfileNames;
+ countfile = validParameter.validFile(parameters, "count", false);
+ if (countfile == "not found") {
+ countfile = "";
+ }else {
+ m->splitAtDash(countfile, countfileNames);
- //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; }
+ //go through files and make sure they are good, if not, then disregard them
+ for (int i = 0; i < countfileNames.size(); i++) {
+
+ bool ignore = false;
+ if (countfileNames[i] == "current") {
+ countfileNames[i] = m->getCountTableFile();
+ if (nameFileNames[i] != "") { m->mothurOut("Using " + countfileNames[i] + " as input file for the count parameter where you had given current."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You have no current count file, ignoring current."); m->mothurOutEndLine(); ignore=true;
+ //erase from file list
+ countfileNames.erase(countfileNames.begin()+i);
+ i--;
+ }
+ }
+
+ if (!ignore) {
+
+ if (inputDir != "") {
+ string path = m->hasPath(countfileNames[i]);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { countfileNames[i] = inputDir + countfileNames[i]; }
+ }
+
+ int ableToOpen;
+ ifstream in;
+
+ ableToOpen = m->openInputFile(countfileNames[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(countfileNames[i]);
+ m->mothurOut("Unable to open " + countfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ countfileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(countfileNames[i]);
+ m->mothurOut("Unable to open " + countfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ countfileNames[i] = tryPath;
+ }
+ }
+
+ in.close();
+
+ if (ableToOpen == 1) {
+ m->mothurOut("Unable to open " + countfileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
+ //erase from file list
+ countfileNames.erase(countfileNames.begin()+i);
+ i--;
+ }else {
+ m->setCountTableFile(countfileNames[i]);
+ }
+ }
+ }
}
-
- 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 (countfileNames.size() != 0) { hasCount = true; }
+
+ //make sure there is at least one valid file left
+ if (hasName && hasCount) { m->mothurOut("[ERROR]: You must enter ONLY ONE of the following: count or name."); m->mothurOutEndLine(); abort = true; }
+
+ if (!hasName && hasCount) { nameFileNames = countfileNames; }
+
+ if ((hasCount || hasName) && (nameFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of name or count files does not match the number of fastafiles, please correct."); m->mothurOutEndLine(); abort=true; }
bool hasGroup = true;
groupfile = validParameter.validFile(parameters, "group", false);
if (hasGroup && (groupFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of groupfiles does not match the number of fastafiles, please correct."); m->mothurOutEndLine(); abort=true; }
+ if (hasGroup && hasCount) { m->mothurOut("[ERROR]: You must enter ONLY ONE of the following: count or group."); 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 = ""; }
+
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
}
}
}else if (hasName) { templatefile = "self"; }
+ else if (hasCount) { templatefile = "self"; }
else {
if (rdb->getSavedReference() != "") {
templatefile = rdb->getSavedReference();
maxp = validParameter.validFile(parameters, "maxp", false); if (maxp == "not found") { useMaxp = false; maxp = "2"; } else{ useMaxp = true; }
minlen = validParameter.validFile(parameters, "minlen", false); if (minlen == "not found") { useMinlen = false; minlen = "10"; } else{ useMinlen = true; }
maxlen = validParameter.validFile(parameters, "maxlen", false); if (maxlen == "not found") { useMaxlen = false; maxlen = "10000"; } else{ useMaxlen = true; }
+
+ strand = validParameter.validFile(parameters, "strand", false); if (strand == "not found") { strand = ""; }
temp = validParameter.validFile(parameters, "ucl", false); if (temp == "not found") { temp = "f"; }
ucl = m->isTrue(temp);
temp = validParameter.validFile(parameters, "skipgaps2", false); if (temp == "not found") { temp = "t"; }
skipgaps2 = m->isTrue(temp);
+
+
+ temp = validParameter.validFile(parameters, "dereplicate", false);
+ if (temp == "not found") { temp = "false"; }
+ dups = m->isTrue(temp);
+
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; }
+ if (hasCount && (templatefile != "self")) { m->mothurOut("You have provided a countfile 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; }
if (hasGroup && (templatefile != "self")) { m->mothurOut("You have provided a group file 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; }
//look for uchime exe
string uchimeCommand;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
uchimeCommand = path + "uchime"; // format the database, -o option gives us the ability
+ if (m->debug) {
+ m->mothurOut("[DEBUG]: Uchime location using \"which uchime\" = ");
+ Command* newCommand = new SystemCommand("which uchime"); m->mothurOutEndLine();
+ newCommand->execute();
+ delete newCommand;
+ m->mothurOut("[DEBUG]: Mothur's location using \"which mothur\" = ");
+ newCommand = new SystemCommand("which mothur"); m->mothurOutEndLine();
+ newCommand->execute();
+ delete newCommand;
+ }
#else
uchimeCommand = path + "uchime.exe";
#endif
-
+
//test to make sure uchime exists
ifstream in;
uchimeCommand = m->getFullPathName(uchimeCommand);
int ableToOpen = m->openInputFile(uchimeCommand, in, "no error"); in.close();
- if(ableToOpen == 1) { m->mothurOut("[ERROR]: " + uchimeCommand + " file does not exist. mothur requires the uchime executable."); m->mothurOutEndLine(); abort = true; }
- }
+ if(ableToOpen == 1) {
+ m->mothurOut(uchimeCommand + " file does not exist. Checking path... \n");
+ //check to see if uchime is in the path??
+
+ string uLocation = m->findProgramPath("uchime");
+
+
+ ifstream in2;
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
+ ableToOpen = m->openInputFile(uLocation, in2, "no error"); in2.close();
+#else
+ ableToOpen = m->openInputFile((uLocation + ".exe"), in2, "no error"); in2.close();
+#endif
+
+ if(ableToOpen == 1) { m->mothurOut("[ERROR]: " + uLocation + " file does not exist. mothur requires the uchime executable."); m->mothurOutEndLine(); abort = true; }
+ else { m->mothurOut("Found uchime in your path, using " + uLocation + "\n");uchimeLocation = uLocation; }
+ }else { uchimeLocation = uchimeCommand; }
+
+ uchimeLocation = m->getFullPathName(uchimeLocation);
+ }
}
catch(exception& e) {
m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
int ChimeraUchimeCommand::execute(){
try{
- if (abort == true) { if (calledHelp) { return 0; } return 2; }
+
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
m->mothurOut("\nuchime by Robert C. Edgar\nhttp://drive5.com/uchime\nThis code is donated to the public domain.\n\n");
int start = time(NULL);
string nameFile = "";
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])) + "uchime.chimera";
- string accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "uchime.accnos";
- string alnsFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "uchime.alns";
+ map<string, string> variables;
+ variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s]));
+ string outputFileName = getOutputFileName("chimera", variables);
+ string accnosFileName = getOutputFileName("accnos", variables);
+ string alnsFileName = getOutputFileName("alns", variables);
string newFasta = m->getRootName(fastaFileNames[s]) + "temp";
+ string newCountFile = "";
//you provided a groupfile
string groupFile = "";
- if (groupFileNames.size() != 0) { groupFile = groupFileNames[s]; }
-
- if ((templatefile == "self") && (groupFile == "")) { //you want to run uchime with a reference template
+ bool hasGroup = false;
+ if (groupFileNames.size() != 0) { groupFile = groupFileNames[s]; hasGroup = true; }
+ else if (hasCount) {
+ CountTable ct;
+ if (ct.testGroups(nameFileNames[s])) { hasGroup = true; }
+ variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(nameFileNames[s]));
+ newCountFile = getOutputFileName("count", variables);
+ }
+
+ if ((templatefile == "self") && (!hasGroup)) { //you want to run uchime with a template=self and no groups
if (processors != 1) { m->mothurOut("When using template=self, mothur can only use 1 processor, continuing."); m->mothurOutEndLine(); processors = 1; }
if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
//read namefile
vector<seqPriorityNode> nameMapCount;
- int error = m->readNames(nameFile, nameMapCount, seqs); if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
+ int error;
+ if (hasCount) {
+ CountTable ct;
+ ct.readTable(nameFile, true, false);
+ for(map<string, string>::iterator it = seqs.begin(); it != seqs.end(); it++) {
+ int num = ct.getNumSeqs(it->first);
+ if (num == 0) { error = 1; }
+ else {
+ seqPriorityNode temp(num, it->second, it->first);
+ nameMapCount.push_back(temp);
+ }
+ }
+ }else {
+ error = m->readNames(nameFile, nameMapCount, seqs); if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
+ }
if (error == 1) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
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(); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
- if (groupFile != "") {
+ if (hasGroup) {
if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
nameFile = nameFileNames[s];
}else { nameFile = getNamesFile(fastaFileNames[s]); }
//Parse sequences by group
- SequenceParser parser(groupFile, fastaFileNames[s], nameFile);
- vector<string> groups = parser.getNamesOfGroups();
+ vector<string> groups;
+ map<string, string> uniqueNames;
+ if (hasCount) {
+ cparser = new SequenceCountParser(nameFile, fastaFileNames[s]);
+ groups = cparser->getNamesOfGroups();
+ uniqueNames = cparser->getAllSeqsMap();
+ }else{
+ sparser = new SequenceParser(groupFile, fastaFileNames[s], nameFile);
+ groups = sparser->getNamesOfGroups();
+ uniqueNames = sparser->getAllSeqsMap();
+ }
if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
if (chimealns) { m->openOutputFile(alnsFileName, out2); out2.close(); }
int totalSeqs = 0;
- if(processors == 1) { totalSeqs = driverGroups(parser, outputFileName, newFasta, accnosFileName, alnsFileName, 0, groups.size(), groups); }
- else { totalSeqs = createProcessesGroups(parser, outputFileName, newFasta, accnosFileName, alnsFileName, groups, nameFile, groupFile, fastaFileNames[s]); }
+ if(processors == 1) { totalSeqs = driverGroups(outputFileName, newFasta, accnosFileName, alnsFileName, newCountFile, 0, groups.size(), groups);
+
+ if (hasCount && dups) {
+ CountTable c; c.readTable(nameFile, true, false);
+ if (!m->isBlank(newCountFile)) {
+ ifstream in2;
+ m->openInputFile(newCountFile, in2);
+
+ string name, group;
+ while (!in2.eof()) {
+ in2 >> name >> group; m->gobble(in2);
+ c.setAbund(name, group, 0);
+ }
+ in2.close();
+ }
+ m->mothurRemove(newCountFile);
+ c.printTable(newCountFile);
+ }
- if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
+ }else { totalSeqs = createProcessesGroups(outputFileName, newFasta, accnosFileName, alnsFileName, newCountFile, groups, nameFile, groupFile, fastaFileNames[s]); }
- int totalChimeras = deconvoluteResults(parser, outputFileName, accnosFileName, alnsFileName);
-
- m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(totalSeqs) + " sequences. " + toString(totalChimeras) + " chimeras were found."); m->mothurOutEndLine();
- m->mothurOut("The number of sequences checked may be larger than the number of unique sequences because some sequences are found in several samples."); m->mothurOutEndLine();
+ if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
+
+
+ if (!dups) {
+ int totalChimeras = deconvoluteResults(uniqueNames, outputFileName, accnosFileName, alnsFileName);
+ m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(totalSeqs) + " sequences. " + toString(totalChimeras) + " chimeras were found."); m->mothurOutEndLine();
+ m->mothurOut("The number of sequences checked may be larger than the number of unique sequences because some sequences are found in several samples."); m->mothurOutEndLine();
+ }else {
+
+ if (hasCount) {
+ set<string> doNotRemove;
+ CountTable c; c.readTable(newCountFile, true, true);
+ vector<string> namesInTable = c.getNamesOfSeqs();
+ for (int i = 0; i < namesInTable.size(); i++) {
+ int temp = c.getNumSeqs(namesInTable[i]);
+ if (temp == 0) { c.remove(namesInTable[i]); }
+ else { doNotRemove.insert((namesInTable[i])); }
+ }
+ //remove names we want to keep from accnos file.
+ set<string> accnosNames = m->readAccnos(accnosFileName);
+ ofstream out2;
+ m->openOutputFile(accnosFileName, out2);
+ for (set<string>::iterator it = accnosNames.begin(); it != accnosNames.end(); it++) {
+ if (doNotRemove.count(*it) == 0) { out2 << (*it) << endl; }
+ }
+ out2.close();
+ c.printTable(newCountFile);
+ outputNames.push_back(newCountFile); outputTypes["count"].push_back(newCountFile);
+ }
+ }
+
+ if (hasCount) { delete cparser; }
+ else { delete sparser; }
+
if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } return 0; }
}else{
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
}
+
+ itTypes = outputTypes.find("count");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setCountTableFile(current); }
+ }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
}
}
//**********************************************************************************************************************
-int ChimeraUchimeCommand::deconvoluteResults(SequenceParser& parser, string outputFileName, string accnosFileName, string alnsFileName){
+int ChimeraUchimeCommand::deconvoluteResults(map<string, string>& uniqueNames, string outputFileName, string accnosFileName, string alnsFileName){
try {
- map<string, string> uniqueNames = parser.getAllSeqsMap();
map<string, string>::iterator itUnique;
int total = 0;
- //edit accnos file
- ifstream in2;
- m->openInputFile(accnosFileName, in2);
-
ofstream out2;
m->openOutputFile(accnosFileName+".temp", out2);
set<string> chimerasInFile;
set<string>::iterator itChimeras;
-
- while (!in2.eof()) {
- if (m->control_pressed) { in2.close(); out2.close(); m->mothurRemove(outputFileName); m->mothurRemove((accnosFileName+".temp")); return 0; }
-
- in2 >> name; m->gobble(in2);
-
- //find unique name
- itUnique = uniqueNames.find(name);
-
- if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing accnos results. Cannot find "+ name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
- else {
- itChimeras = chimerasInFile.find((itUnique->second));
-
- if (itChimeras == chimerasInFile.end()) {
- out2 << itUnique->second << endl;
- chimerasInFile.insert((itUnique->second));
- total++;
- }
- }
- }
- in2.close();
+ if (!m->isBlank(accnosFileName)) {
+ //edit accnos file
+ ifstream in2;
+ m->openInputFile(accnosFileName, in2);
+
+ while (!in2.eof()) {
+ if (m->control_pressed) { in2.close(); out2.close(); m->mothurRemove(outputFileName); m->mothurRemove((accnosFileName+".temp")); return 0; }
+
+ in2 >> name; m->gobble(in2);
+
+ //find unique name
+ itUnique = uniqueNames.find(name);
+
+ if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing accnos results. Cannot find " + name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
+ else {
+ itChimeras = chimerasInFile.find((itUnique->second));
+
+ if (itChimeras == chimerasInFile.end()) {
+ out2 << itUnique->second << endl;
+ chimerasInFile.insert((itUnique->second));
+ total++;
+ }
+ }
+ }
+ in2.close();
+ }
out2.close();
m->mothurRemove(accnosFileName);
}
}
//**********************************************************************************************************************
-int ChimeraUchimeCommand::driverGroups(SequenceParser& parser, string outputFName, string filename, string accnos, string alns, int start, int end, vector<string> groups){
+int ChimeraUchimeCommand::driverGroups(string outputFName, string filename, string accnos, string alns, string countlist, int start, int end, vector<string> groups){
try {
int totalSeqs = 0;
int numChimeras = 0;
-
+
+
+ ofstream outCountList;
+ if (hasCount && dups) { m->openOutputFile(countlist, outCountList); }
+
for (int i = start; i < end; i++) {
- int start = time(NULL); if (m->control_pressed) { return 0; }
+ int start = time(NULL); if (m->control_pressed) { outCountList.close(); m->mothurRemove(countlist); return 0; }
+
+ int error;
+ if (hasCount) { error = cparser->getSeqs(groups[i], filename, true); if ((error == 1) || m->control_pressed) { return 0; } }
+ else { error = sparser->getSeqs(groups[i], filename, true); if ((error == 1) || m->control_pressed) { return 0; } }
- int error = parser.getSeqs(groups[i], filename, true); if ((error == 1) || m->control_pressed) { return 0; }
-
- int numSeqs = driver((outputFName + groups[i]), filename, (accnos+ groups[i]), (alns+ groups[i]), numChimeras);
+ int numSeqs = driver((outputFName + groups[i]), filename, (accnos+groups[i]), (alns+ groups[i]), numChimeras);
totalSeqs += numSeqs;
if (m->control_pressed) { return 0; }
//remove file made for uchime
- m->mothurRemove(filename);
-
+ if (!m->debug) { m->mothurRemove(filename); }
+ else { m->mothurOut("[DEBUG]: saving file: " + filename + ".\n"); }
+
+ //if we provided a count file with group info and set dereplicate=t, then we want to create a *.pick.count_table
+ //This table will zero out group counts for seqs determined to be chimeric by that group.
+ if (dups) {
+ if (!m->isBlank(accnos+groups[i])) {
+ ifstream in;
+ m->openInputFile(accnos+groups[i], in);
+ string name;
+ if (hasCount) {
+ while (!in.eof()) {
+ in >> name; m->gobble(in);
+ outCountList << name << '\t' << groups[i] << endl;
+ }
+ in.close();
+ }else {
+ map<string, string> thisnamemap = sparser->getNameMap(groups[i]);
+ map<string, string>::iterator itN;
+ ofstream out;
+ m->openOutputFile(accnos+groups[i]+".temp", out);
+ while (!in.eof()) {
+ in >> name; m->gobble(in);
+ itN = thisnamemap.find(name);
+ if (itN != thisnamemap.end()) {
+ vector<string> tempNames; m->splitAtComma(itN->second, tempNames);
+ for (int j = 0; j < tempNames.size(); j++) { out << tempNames[j] << endl; }
+
+ }else { m->mothurOut("[ERROR]: parsing cannot find " + name + ".\n"); m->control_pressed = true; }
+ }
+ out.close();
+ in.close();
+ m->renameFile(accnos+groups[i]+".temp", accnos+groups[i]);
+ }
+
+ }
+ }
+
//append files
m->appendFiles((outputFName+groups[i]), outputFName); m->mothurRemove((outputFName+groups[i]));
m->appendFiles((accnos+groups[i]), accnos); m->mothurRemove((accnos+groups[i]));
if (chimealns) { m->appendFiles((alns+groups[i]), alns); m->mothurRemove((alns+groups[i])); }
m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences from group " + groups[i] + "."); m->mothurOutEndLine();
- }
-
- return totalSeqs;
+ }
+
+ if (hasCount && dups) { outCountList.close(); }
+
+ return totalSeqs;
}
catch(exception& e) {
vector<char*> cPara;
- string path = m->argv;
- string tempPath = path;
- for (int i = 0; i < path.length(); i++) { tempPath[i] = tolower(path[i]); }
- path = path.substr(0, (tempPath.find_last_of('m')));
-
- string uchimeCommand = path;
-#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
- uchimeCommand += "uchime ";
-#else
- uchimeCommand += "uchime";
- uchimeCommand = "\"" + uchimeCommand + "\"";
-#endif
-
- char* tempUchime;
+ string uchimeCommand = uchimeLocation;
+ uchimeCommand = "\"" + uchimeCommand + "\" ";
+
+ char* tempUchime;
tempUchime= new char[uchimeCommand.length()+1];
*tempUchime = '\0';
strncat(tempUchime, uchimeCommand.c_str(), uchimeCommand.length());
cPara.push_back(tempUchime);
- char* tempIn = new char[8];
- *tempIn = '\0'; strncat(tempIn, "--input", 7);
- //strcpy(tempIn, "--input");
- cPara.push_back(tempIn);
- char* temp = new char[filename.length()+1];
- *temp = '\0'; strncat(temp, filename.c_str(), filename.length());
- //strcpy(temp, filename.c_str());
- cPara.push_back(temp);
-
- //are you using a reference file
+ //are you using a reference file
if (templatefile != "self") {
+ string outputFileName = filename.substr(1, filename.length()-2) + ".uchime_formatted";
+ prepFile(filename.substr(1, filename.length()-2), outputFileName);
+ filename = outputFileName;
+ filename = "\"" + filename + "\"";
//add reference file
char* tempRef = new char[5];
//strcpy(tempRef, "--db");
cPara.push_back(tempR);
}
+ char* tempIn = new char[8];
+ *tempIn = '\0'; strncat(tempIn, "--input", 7);
+ //strcpy(tempIn, "--input");
+ cPara.push_back(tempIn);
+ char* temp = new char[filename.length()+1];
+ *temp = '\0'; strncat(temp, filename.c_str(), filename.length());
+ //strcpy(temp, filename.c_str());
+ cPara.push_back(temp);
+
char* tempO = new char[12];
*tempO = '\0'; strncat(tempO, "--uchimeout", 11);
//strcpy(tempO, "--uchimeout");
*tempa = '\0'; strncat(tempa, alns.c_str(), alns.length());
cPara.push_back(tempa);
}
+
+ if (strand != "") {
+ char* tempA = new char[9];
+ *tempA = '\0'; strncat(tempA, "--strand", 8);
+ cPara.push_back(tempA);
+ char* tempa = new char[strand.length()+1];
+ *tempa = '\0'; strncat(tempa, strand.c_str(), strand.length());
+ cPara.push_back(tempa);
+ }
if (useAbskew) {
char* tempskew = new char[9];
#else
commandString = "\"" + commandString + "\"";
#endif
+ if (m->debug) { m->mothurOut("[DEBUG]: uchime command = " + commandString + ".\n"); }
system(commandString.c_str());
//free memory
string name = "";
string chimeraFlag = "";
- in >> chimeraFlag >> name;
-
- //fix name if needed
- if (templatefile == "self") {
- name = name.substr(0, name.length()-1); //rip off last /
- name = name.substr(0, name.find_last_of('/'));
+ //in >> chimeraFlag >> name;
+
+ string line = m->getline(in);
+ vector<string> pieces = m->splitWhiteSpace(line);
+ if (pieces.size() > 2) {
+ name = pieces[1];
+ //fix name if needed
+ if (templatefile == "self") {
+ name = name.substr(0, name.length()-1); //rip off last /
+ name = name.substr(0, name.find_last_of('/'));
+ }
+
+ chimeraFlag = pieces[pieces.size()-1];
}
-
- for (int i = 0; i < 15; i++) { in >> chimeraFlag; }
+ //for (int i = 0; i < 15; i++) { in >> chimeraFlag; }
m->gobble(in);
if (chimeraFlag == "Y") { out << name << endl; numChimeras++; }
in.close();
out.close();
+ //if (templatefile != "self") { m->mothurRemove(filename); }
+
return num;
}
catch(exception& e) {
}
}
/**************************************************************************************************/
+//uchime can't handle some of the things allowed in mothurs fasta files. This functions "cleans up" the file.
+int ChimeraUchimeCommand::prepFile(string filename, string output) {
+ try {
+
+ ifstream in;
+ m->openInputFile(filename, in);
+
+ ofstream out;
+ m->openOutputFile(output, out);
+
+ while (!in.eof()) {
+ if (m->control_pressed) { break; }
+
+ Sequence seq(in); m->gobble(in);
+
+ if (seq.getName() != "") { seq.printSequence(out); }
+ }
+ in.close();
+ out.close();
+
+ return 0;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraUchimeCommand", "prepFile");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
int ChimeraUchimeCommand::createProcesses(string outputFileName, string filename, string accnos, string alns, int& numChimeras) {
try {
// Allocate memory for thread data.
string extension = toString(i) + ".temp";
- uchimeData* tempUchime = new uchimeData(outputFileName+extension, templatefile, files[i], "", "", "", accnos+extension, alns+extension, dummy, m, 0, 0, i);
- tempUchime->setBooleans(useAbskew, chimealns, useMinH, useMindiv, useXn, useDn, useXa, useChunks, useMinchunk, useIdsmoothwindow, useMinsmoothid, useMaxp, skipgaps, skipgaps2, useMinlen, useMaxlen, ucl, useQueryfract);
- tempUchime->setVariables(abskew, minh, mindiv, xn, dn, xa, chunks, minchunk, idsmoothwindow, minsmoothid, maxp, minlen, maxlen, queryfract);
+ uchimeData* tempUchime = new uchimeData(outputFileName+extension, uchimeLocation, templatefile, files[i], "", "", "", accnos+extension, alns+extension, "", dummy, m, 0, 0, i);
+ tempUchime->setBooleans(dups, useAbskew, chimealns, useMinH, useMindiv, useXn, useDn, useXa, useChunks, useMinchunk, useIdsmoothwindow, useMinsmoothid, useMaxp, skipgaps, skipgaps2, useMinlen, useMaxlen, ucl, useQueryfract, hasCount);
+ tempUchime->setVariables(abskew, minh, mindiv, xn, dn, xa, chunks, minchunk, idsmoothwindow, minsmoothid, maxp, minlen, maxlen, queryfract, strand);
pDataArray.push_back(tempUchime);
processIDS.push_back(i);
}
/**************************************************************************************************/
-int ChimeraUchimeCommand::createProcessesGroups(SequenceParser& parser, string outputFName, string filename, string accnos, string alns, vector<string> groups, string nameFile, string groupFile, string fastaFile) {
+int ChimeraUchimeCommand::createProcessesGroups(string outputFName, string filename, string accnos, string alns, string newCountFile, vector<string> groups, string nameFile, string groupFile, string fastaFile) {
try {
processIDS.clear();
int process = 1;
int num = 0;
+
+ CountTable newCount;
+ if (hasCount && dups) { newCount.readTable(nameFile, true, false); }
//sanity check
if (groups.size() < processors) { processors = groups.size(); }
//divide the groups between the processors
vector<linePair> lines;
- int numGroupsPerProcessor = groups.size() / processors;
- for (int i = 0; i < processors; i++) {
- int startIndex = i * numGroupsPerProcessor;
- int endIndex = (i+1) * numGroupsPerProcessor;
- if(i == (processors - 1)){ endIndex = groups.size(); }
- lines.push_back(linePair(startIndex, endIndex));
- }
-
+ int remainingPairs = groups.size();
+ int startIndex = 0;
+ for (int remainingProcessors = processors; remainingProcessors > 0; remainingProcessors--) {
+ int numPairs = remainingPairs; //case for last processor
+ if (remainingProcessors != 1) { numPairs = ceil(remainingPairs / remainingProcessors); }
+ lines.push_back(linePair(startIndex, (startIndex+numPairs))); //startIndex, endIndex
+ startIndex = startIndex + numPairs;
+ remainingPairs = remainingPairs - numPairs;
+ }
+
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
//loop through and create all the processes you want
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 = driverGroups(parser, outputFName + toString(getpid()) + ".temp", filename + toString(getpid()) + ".temp", accnos + toString(getpid()) + ".temp", alns + toString(getpid()) + ".temp", lines[process].start, lines[process].end, groups);
+ num = driverGroups(outputFName + toString(getpid()) + ".temp", filename + toString(getpid()) + ".temp", accnos + toString(getpid()) + ".temp", alns + toString(getpid()) + ".temp", accnos + ".byCount." + toString(getpid()) + ".temp", lines[process].start, lines[process].end, groups);
//pass numSeqs to parent
ofstream out;
}
//do my part
- num = driverGroups(parser, outputFName, filename, accnos, alns, lines[0].start, lines[0].end, groups);
+ num = driverGroups(outputFName, filename, accnos, alns, accnos + ".byCount", lines[0].start, lines[0].end, groups);
//force parent to wait until all the processes are done
for (int i=0;i<processIDS.size();i++) {
int temp = processIDS[i];
wait(&temp);
}
-
+
for (int i = 0; i < processIDS.size(); i++) {
ifstream in;
string tempFile = outputFName + toString(processIDS[i]) + ".num.temp";
m->openInputFile(tempFile, in);
if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
in.close(); m->mothurRemove(tempFile);
- }
-
+ }
+
#else
//////////////////////////////////////////////////////////////////////////////////////////////////////
//Windows version shared memory, so be careful when passing variables through the uchimeData struct.
// Allocate memory for thread data.
string extension = toString(i) + ".temp";
- uchimeData* tempUchime = new uchimeData(outputFName+extension, templatefile, filename+extension, fastaFile, nameFile, groupFile, accnos+extension, alns+extension, groups, m, lines[i].start, lines[i].end, i);
- tempUchime->setBooleans(useAbskew, chimealns, useMinH, useMindiv, useXn, useDn, useXa, useChunks, useMinchunk, useIdsmoothwindow, useMinsmoothid, useMaxp, skipgaps, skipgaps2, useMinlen, useMaxlen, ucl, useQueryfract);
- tempUchime->setVariables(abskew, minh, mindiv, xn, dn, xa, chunks, minchunk, idsmoothwindow, minsmoothid, maxp, minlen, maxlen, queryfract);
+ uchimeData* tempUchime = new uchimeData(outputFName+extension, uchimeLocation, templatefile, filename+extension, fastaFile, nameFile, groupFile, accnos+extension, alns+extension, accnos+".byCount."+extension, groups, m, lines[i].start, lines[i].end, i);
+ tempUchime->setBooleans(dups, useAbskew, chimealns, useMinH, useMindiv, useXn, useDn, useXa, useChunks, useMinchunk, useIdsmoothwindow, useMinsmoothid, useMaxp, skipgaps, skipgaps2, useMinlen, useMaxlen, ucl, useQueryfract, hasCount);
+ tempUchime->setVariables(abskew, minh, mindiv, xn, dn, xa, chunks, minchunk, idsmoothwindow, minsmoothid, maxp, minlen, maxlen, queryfract, strand);
pDataArray.push_back(tempUchime);
processIDS.push_back(i);
//using the main process as a worker saves time and memory
- num = driverGroups(parser, outputFName, filename, accnos, alns, lines[0].start, lines[0].end, groups);
+ num = driverGroups(outputFName, filename, accnos, alns, accnos + ".byCount", lines[0].start, lines[0].end, groups);
//Wait until all threads have terminated.
WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE);
CloseHandle(hThreadArray[i]);
delete pDataArray[i];
}
+
+
#endif
-
-
+
+ //read my own
+ if (hasCount && dups) {
+ if (!m->isBlank(accnos + ".byCount")) {
+ ifstream in2;
+ m->openInputFile(accnos + ".byCount", in2);
+
+ string name, group;
+ while (!in2.eof()) {
+ in2 >> name >> group; m->gobble(in2);
+ newCount.setAbund(name, group, 0);
+ }
+ in2.close();
+ }
+ m->mothurRemove(accnos + ".byCount");
+ }
+
//append output files
for(int i=0;i<processIDS.size();i++){
m->appendFiles((outputFName + toString(processIDS[i]) + ".temp"), outputFName);
m->appendFiles((alns + toString(processIDS[i]) + ".temp"), alns);
m->mothurRemove((alns + toString(processIDS[i]) + ".temp"));
}
+
+ if (hasCount && dups) {
+ if (!m->isBlank(accnos + ".byCount." + toString(processIDS[i]) + ".temp")) {
+ ifstream in2;
+ m->openInputFile(accnos + ".byCount." + toString(processIDS[i]) + ".temp", in2);
+
+ string name, group;
+ while (!in2.eof()) {
+ in2 >> name >> group; m->gobble(in2);
+ newCount.setAbund(name, group, 0);
+ }
+ in2.close();
+ }
+ m->mothurRemove(accnos + ".byCount." + toString(processIDS[i]) + ".temp");
+ }
+
}
-
+
+ //print new *.pick.count_table
+ if (hasCount && dups) { newCount.printTable(newCountFile); }
+
return num;
}
projects / mothur.git / blobdiff