*/
#include "trimseqscommand.h"
+#include "needlemanoverlap.hpp"
//***************************************************************************************************************
-TrimSeqsCommand::TrimSeqsCommand(string option){
+TrimSeqsCommand::TrimSeqsCommand(string option) {
try {
abort = false;
else {
//valid paramters for this command
- string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile", "qthreshold", "qaverage", "allfiles"};
+ string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
+ "qthreshold", "qaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
map<string,string> parameters = parser.getParameters();
ValidParameters validParameter;
+ map<string,string>::iterator it;
//check to make sure all parameters are valid for command
- for (map<string,string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
+ for (it = parameters.begin(); it != parameters.end(); it++) {
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ //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("fasta");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["fasta"] = inputDir + it->second; }
+ }
+
+ it = parameters.find("oligos");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["oligos"] = inputDir + it->second; }
+ }
+
+ it = parameters.find("qfile");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["qfile"] = inputDir + it->second; }
+ }
+ }
+
+
//check for required parameters
fastaFile = validParameter.validFile(parameters, "fasta", true);
- if (fastaFile == "not found") { mothurOut("fasta is a required parameter for the screen.seqs command."); mothurOutEndLine(); abort = true; }
+ if (fastaFile == "not found") { m->mothurOut("fasta is a required parameter for the screen.seqs command."); m->mothurOutEndLine(); abort = true; }
else if (fastaFile == "not open") { abort = true; }
-
+
+ //if the user changes the output directory command factory will send this info to us in the output parameter
+ outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
+ outputDir = "";
+ outputDir += hasPath(fastaFile); //if user entered a file with a path then preserve it
+ }
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
temp = validParameter.validFile(parameters, "maxlength", false); if (temp == "not found") { temp = "0"; }
convert(temp, maxLength);
+ temp = validParameter.validFile(parameters, "tdiffs", false); if (temp == "not found") { temp = "0"; }
+ convert(temp, tdiffs);
+
+ temp = validParameter.validFile(parameters, "bdiffs", false); if (temp == "not found") { temp = "0"; }
+ convert(temp, bdiffs);
+
+ temp = validParameter.validFile(parameters, "pdiffs", false); if (temp == "not found") { temp = "0"; }
+ convert(temp, pdiffs);
+
temp = validParameter.validFile(parameters, "qfile", true);
if (temp == "not found") { qFileName = ""; }
- else if(temp == "not open") { abort = 0; }
+ else if(temp == "not open") { abort = true; }
else { qFileName = temp; }
temp = validParameter.validFile(parameters, "qthreshold", false); if (temp == "not found") { temp = "0"; }
convert(temp, qThreshold);
+
+ temp = validParameter.validFile(parameters, "qtrim", false); if (temp == "not found") { temp = "F"; }
+ qtrim = isTrue(temp);
temp = validParameter.validFile(parameters, "qaverage", false); if (temp == "not found") { temp = "0"; }
convert(temp, qAverage);
temp = validParameter.validFile(parameters, "allfiles", false); if (temp == "not found") { temp = "F"; }
allFiles = isTrue(temp);
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
+ convert(temp, processors);
+
if(allFiles && oligoFile == ""){
- mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); mothurOutEndLine();
+ m->mothurOut("You selected allfiles, but didn't enter an oligos file. Ignoring the allfiles request."); m->mothurOutEndLine();
}
if((qAverage != 0 && qThreshold != 0) && qFileName == ""){
- mothurOut("You didn't provide a quality file name, quality criteria will be ignored."); mothurOutEndLine();
+ m->mothurOut("You didn't provide a quality file name, quality criteria will be ignored."); m->mothurOutEndLine();
qAverage=0;
qThreshold=0;
}
if(!flip && oligoFile=="" && !maxLength && !minLength && (maxAmbig==-1) && !maxHomoP && qFileName == ""){
- mothurOut("You didn't set any options... quiting command."); mothurOutEndLine();
+ m->mothurOut("You didn't set any options... quiting command."); m->mothurOutEndLine();
abort = true;
}
}
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "TrimSeqsCommand");
+ m->errorOut(e, "TrimSeqsCommand", "TrimSeqsCommand");
exit(1);
}
}
void TrimSeqsCommand::help(){
try {
- mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
- mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength and maxlength.\n");
- mothurOut("The fasta parameter is required.\n");
- mothurOut("The flip parameter .... The default is 0.\n");
- mothurOut("The oligos parameter .... The default is "".\n");
- mothurOut("The maxambig parameter .... The default is -1.\n");
- mothurOut("The maxhomop parameter .... The default is 0.\n");
- mothurOut("The minlength parameter .... The default is 0.\n");
- mothurOut("The maxlength parameter .... The default is 0.\n");
- mothurOut("The trim.seqs command should be in the following format: \n");
- mothurOut("trim.seqs(fasta=yourFastaFile, flip=yourFlip, oligos=yourOligos, maxambig=yourMaxambig, \n");
- mothurOut("maxhomop=yourMaxhomop, minlength=youMinlength, maxlength=yourMaxlength) \n");
- mothurOut("Example trim.seqs(fasta=abrecovery.fasta, flip=..., oligos=..., maxambig=..., maxhomop=..., minlength=..., maxlength=...).\n");
- mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n");
+ m->mothurOut("The trim.seqs command reads a fastaFile and creates .....\n");
+ m->mothurOut("The trim.seqs command parameters are fasta, flip, oligos, maxambig, maxhomop, minlength, maxlength, qfile, qthreshold, qaverage, diffs, qtrim and allfiles.\n");
+ m->mothurOut("The fasta parameter is required.\n");
+ m->mothurOut("The flip parameter .... The default is 0.\n");
+ m->mothurOut("The oligos parameter .... The default is "".\n");
+ m->mothurOut("The maxambig parameter .... The default is -1.\n");
+ m->mothurOut("The maxhomop parameter .... The default is 0.\n");
+ m->mothurOut("The minlength parameter .... The default is 0.\n");
+ m->mothurOut("The maxlength parameter .... The default is 0.\n");
+ m->mothurOut("The tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is 0.\n");
+ m->mothurOut("The bdiffs parameter is used to specify the number of differences allowed in the barcode. The default is 0.\n");
+ m->mothurOut("The pdiffs parameter is used to specify the number of differences allowed in the primer. The default is 0.\n");
+ m->mothurOut("The qfile parameter .....\n");
+ m->mothurOut("The qthreshold parameter .... The default is 0.\n");
+ m->mothurOut("The qaverage parameter .... The default is 0.\n");
+ m->mothurOut("The allfiles parameter .... The default is F.\n");
+ m->mothurOut("The qtrim parameter .... The default is F.\n");
+ m->mothurOut("The trim.seqs command should be in the following format: \n");
+ m->mothurOut("trim.seqs(fasta=yourFastaFile, flip=yourFlip, oligos=yourOligos, maxambig=yourMaxambig, \n");
+ m->mothurOut("maxhomop=yourMaxhomop, minlength=youMinlength, maxlength=yourMaxlength) \n");
+ m->mothurOut("Example trim.seqs(fasta=abrecovery.fasta, flip=..., oligos=..., maxambig=..., maxhomop=..., minlength=..., maxlength=...).\n");
+ m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n");
+ m->mothurOut("For more details please check out the wiki http://www.mothur.org/wiki/Trim.seqs .\n\n");
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "help");
+ m->errorOut(e, "TrimSeqsCommand", "help");
exit(1);
}
}
try{
if (abort == true) { return 0; }
+
+ numFPrimers = 0; //this needs to be initialized
+ numRPrimers = 0;
+
+ string trimSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "trim.fasta";
+ outputNames.push_back(trimSeqFile);
+ string scrapSeqFile = outputDir + getRootName(getSimpleName(fastaFile)) + "scrap.fasta";
+ outputNames.push_back(scrapSeqFile);
+ string groupFile = outputDir + getRootName(getSimpleName(fastaFile)) + "groups";
+
+ vector<string> fastaFileNames;
+ if(oligoFile != ""){
+ outputNames.push_back(groupFile);
+ getOligos(fastaFileNames);
+ }
+
+ if(qFileName != "") { setLines(qFileName, qLines); }
- ifstream inFASTA;
- openInputFile(fastaFile, inFASTA);
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors == 1){
+ ifstream inFASTA;
+ openInputFile(fastaFile, inFASTA);
+ int numSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
+ inFASTA.close();
+
+ lines.push_back(new linePair(0, numSeqs));
+
+ driverCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
+
+ for (int j = 0; j < fastaFileNames.size(); j++) {
+ rename((fastaFileNames[j] + toString(getpid()) + ".temp").c_str(), fastaFileNames[j].c_str());
+ }
+
+ }else{
+ setLines(fastaFile, lines);
+ if(qFileName == "") { qLines = lines; }
+
+ createProcessesCreateTrim(fastaFile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames);
+
+ rename((trimSeqFile + toString(processIDS[0]) + ".temp").c_str(), trimSeqFile.c_str());
+ rename((scrapSeqFile + toString(processIDS[0]) + ".temp").c_str(), scrapSeqFile.c_str());
+ rename((groupFile + toString(processIDS[0]) + ".temp").c_str(), groupFile.c_str());
+ for (int j = 0; j < fastaFileNames.size(); j++) {
+ rename((fastaFileNames[j] + toString(processIDS[0]) + ".temp").c_str(), fastaFileNames[j].c_str());
+ }
+ //append files
+ for(int i=1;i<processors;i++){
+ appendFiles((trimSeqFile + toString(processIDS[i]) + ".temp"), trimSeqFile);
+ remove((trimSeqFile + toString(processIDS[i]) + ".temp").c_str());
+ appendFiles((scrapSeqFile + toString(processIDS[i]) + ".temp"), scrapSeqFile);
+ remove((scrapSeqFile + toString(processIDS[i]) + ".temp").c_str());
+ appendFiles((groupFile + toString(processIDS[i]) + ".temp"), groupFile);
+ remove((groupFile + toString(processIDS[i]) + ".temp").c_str());
+ for (int j = 0; j < fastaFileNames.size(); j++) {
+ appendFiles((fastaFileNames[j] + toString(processIDS[i]) + ".temp"), fastaFileNames[j]);
+ remove((fastaFileNames[j] + toString(processIDS[i]) + ".temp").c_str());
+ }
+ }
+ }
+
+ if (m->control_pressed) { return 0; }
+ #else
+ ifstream inFASTA;
+ openInputFile(fastafileNames[s], inFASTA);
+ numSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
+ inFASTA.close();
+
+ lines.push_back(new linePair(0, numSeqs));
+
+ driverCreateSummary(fastafile, qFileName, trimSeqFile, scrapSeqFile, groupFile, fastaFileNames, lines[0], lines[0]);
+
+ if (m->control_pressed) { return 0; }
+ #endif
+
+
+ for(int i=0;i<fastaFileNames.size();i++){
+ ifstream inFASTA;
+ string seqName;
+ openInputFile(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA);
+ ofstream outGroups;
+ openOutputFile(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups", outGroups);
+ outputNames.push_back(outputDir + getRootName(getSimpleName(fastaFile)) + groupVector[i] + ".groups");
+
+ while(!inFASTA.eof()){
+ if(inFASTA.get() == '>'){
+ inFASTA >> seqName;
+ outGroups << seqName << '\t' << groupVector[i] << endl;
+ }
+ while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
+ }
+ outGroups.close();
+ inFASTA.close();
+ }
+
+ if (m->control_pressed) {
+ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
+ return 0;
+ }
+
+ m->mothurOutEndLine();
+ m->mothurOut("Output File Names: "); m->mothurOutEndLine();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
+ m->mothurOutEndLine();
+
+ return 0;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "execute");
+ exit(1);
+ }
+}
+
+/**************************************************************************************/
+int TrimSeqsCommand::driverCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames, linePair* line, linePair* qline) {
+ try {
ofstream outFASTA;
- string trimSeqFile = getRootName(fastaFile) + "trim.fasta";
- openOutputFile(trimSeqFile, outFASTA);
+ int able = openOutputFile(trimFile, outFASTA);
+
+ ofstream scrapFASTA;
+ openOutputFile(scrapFile, scrapFASTA);
ofstream outGroups;
vector<ofstream*> fastaFileNames;
- if(oligoFile != ""){
- string groupFile = getRootName(fastaFile) + "groups";
- openOutputFile(groupFile, outGroups);
- getOligos(fastaFileNames);
+ if (oligoFile != "") {
+ openOutputFile(groupFile, outGroups);
+ for (int i = 0; i < fastaNames.size(); i++) {
+ fastaFileNames.push_back(new ofstream((fastaNames[i] + toString(getpid()) + ".temp").c_str(), ios::ate));
+ }
}
- ofstream scrapFASTA;
- string scrapSeqFile = getRootName(fastaFile) + "scrap.fasta";
- openOutputFile(scrapSeqFile, scrapFASTA);
+ ifstream inFASTA;
+ openInputFile(filename, inFASTA);
ifstream qFile;
if(qFileName != "") { openInputFile(qFileName, qFile); }
- bool success;
+ qFile.seekg(qline->start);
+ inFASTA.seekg(line->start);
+
+ for(int i=0;i<line->num;i++){
+
+ if (m->control_pressed) {
+ inFASTA.close();
+ outFASTA.close();
+ scrapFASTA.close();
+ if (oligoFile != "") { outGroups.close(); }
+ if(qFileName != "") { qFile.close(); }
+ for(int i=0;i<fastaFileNames.size();i++){
+ fastaFileNames[i]->close();
+ delete fastaFileNames[i];
+ }
+ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
+ return 0;
+ }
+
+ bool success = 1;
- while(!inFASTA.eof()){
Sequence currSeq(inFASTA);
+
string origSeq = currSeq.getUnaligned();
- int group;
- string trashCode = "";
-
- if(qFileName != ""){
- if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); }
- else if(qAverage != 0) { success = cullQualAverage(currSeq, qFile); }
- if(!success) { trashCode += 'q'; }
- }
- if(barcodes.size() != 0){
-
- success = stripBarcode(currSeq, group);
- if(!success){ trashCode += 'b'; }
- }
- if(numFPrimers != 0){
- success = stripForward(currSeq);
- if(!success){ trashCode += 'f'; }
- }
- if(numRPrimers != 0){
- success = stripReverse(currSeq);
- if(!success){ trashCode += 'r'; }
- }
- if(minLength > 0 || maxLength > 0){
- success = cullLength(currSeq);
- if ((currSeq.getUnaligned().length() > 300) && (success)) { cout << "too long " << currSeq.getUnaligned().length() << endl; }
- if(!success){ trashCode += 'l'; }
- }
- if(maxHomoP > 0){
- success = cullHomoP(currSeq);
- if(!success){ trashCode += 'h'; }
- }
- if(maxAmbig != -1){
- success = cullAmbigs(currSeq);
- if(!success){ trashCode += 'n'; }
- }
-
- if(flip){ currSeq.reverseComplement(); } // should go last
+ if (origSeq != "") {
+ int group;
+ string trashCode = "";
+ int currentSeqsDiffs = 0;
+
+ if(qFileName != ""){
+ if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); }
+ else if(qAverage != 0) { success = cullQualAverage(currSeq, qFile); }
+ if ((!qtrim) && (origSeq.length() != currSeq.getUnaligned().length())) {
+ success = 0; //if you don't want to trim and the sequence does not meet quality requirements, move to scrap
+ }
+ if(!success) { trashCode += 'q'; }
+ }
- if(trashCode.length() == 0){
- currSeq.setAligned(currSeq.getUnaligned()); //this is because of a modification we made to the sequence class to fix a bug. all seqs have an aligned version, which is the version that gets printed.
- currSeq.printSequence(outFASTA);
if(barcodes.size() != 0){
- outGroups << currSeq.getName() << '\t' << groupVector[group] << endl;
-
- if(allFiles){
- currSeq.printSequence(*fastaFileNames[group]);
+ success = stripBarcode(currSeq, group);
+ if(!success){ trashCode += 'b'; }
+ else{ currentSeqsDiffs += currentSeqsTdiffs; }
+ }
+
+ if(numFPrimers != 0){
+ success = stripForward(currSeq);
+ if(!success){ trashCode += 'f'; }
+ else{ currentSeqsDiffs += currentSeqsTdiffs; }
+ }
+
+ if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
+
+ if(numRPrimers != 0){
+ success = stripReverse(currSeq);
+ if(!success){ trashCode += 'r'; }
+ }
+
+ if(minLength > 0 || maxLength > 0){
+ success = cullLength(currSeq);
+ if(!success){ trashCode += 'l'; }
+ }
+ if(maxHomoP > 0){
+ success = cullHomoP(currSeq);
+ if(!success){ trashCode += 'h'; }
+ }
+ if(maxAmbig != -1){
+ success = cullAmbigs(currSeq);
+ if(!success){ trashCode += 'n'; }
+ }
+
+ if(flip){ currSeq.reverseComplement(); } // should go last
+
+ if(trashCode.length() == 0){
+ currSeq.setAligned(currSeq.getUnaligned()); //this is because of a modification we made to the sequence class to fix a bug. all seqs have an aligned version, which is the version that gets printed.
+ currSeq.printSequence(outFASTA);
+ if(barcodes.size() != 0){
+ outGroups << currSeq.getName() << '\t' << groupVector[group] << endl;
+
+ if(allFiles){
+ currSeq.printSequence(*fastaFileNames[group]);
+ }
}
}
- }
- else{
- currSeq.setName(currSeq.getName() + '|' + trashCode);
- currSeq.setUnaligned(origSeq);
- currSeq.printSequence(scrapFASTA);
+ else{
+ currSeq.setName(currSeq.getName() + '|' + trashCode);
+ currSeq.setUnaligned(origSeq);
+ currSeq.printSequence(scrapFASTA);
+ }
}
gobble(inFASTA);
}
+
inFASTA.close();
outFASTA.close();
scrapFASTA.close();
- outGroups.close();
+ if (oligoFile != "") { outGroups.close(); }
if(qFileName != "") { qFile.close(); }
for(int i=0;i<fastaFileNames.size();i++){
delete fastaFileNames[i];
}
- for(int i=0;i<fastaFileNames.size();i++){
- string seqName;
- openInputFile(getRootName(fastaFile) + groupVector[i] + ".fasta", inFASTA);
- ofstream outGroups;
- openOutputFile(getRootName(fastaFile) + groupVector[i] + ".groups", outGroups);
+ return 0;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "driverCreateTrim");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+int TrimSeqsCommand::createProcessesCreateTrim(string filename, string qFileName, string trimFile, string scrapFile, string groupFile, vector<string> fastaNames) {
+ try {
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ int process = 0;
+ int exitCommand = 1;
+ processIDS.clear();
+
+ //loop through and create all the processes you want
+ while (process != processors) {
+ int pid = fork();
- while(!inFASTA.eof()){
- if(inFASTA.get() == '>'){
- inFASTA >> seqName;
- outGroups << seqName << '\t' << groupVector[i] << endl;
- }
- while (!inFASTA.eof()) { char c = inFASTA.get(); if (c == 10 || c == 13){ break; } }
- }
- outGroups.close();
- inFASTA.close();
+ if (pid > 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){
+ driverCreateTrim(filename, qFileName, (trimFile + toString(getpid()) + ".temp"), (scrapFile + toString(getpid()) + ".temp"), (groupFile + toString(getpid()) + ".temp"), fastaNames, lines[process], qLines[process]);
+ exit(0);
+ }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
}
+ //force parent to wait until all the processes are done
+ for (int i=0;i<processors;i++) {
+ int temp = processIDS[i];
+ wait(&temp);
+ }
- return 0;
+ return exitCommand;
+#endif
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "execute");
+ m->errorOut(e, "TrimSeqsCommand", "createProcessesCreateTrim");
exit(1);
}
}
+/**************************************************************************************************/
+int TrimSeqsCommand::setLines(string filename, vector<linePair*>& lines) {
+ try {
+
+ lines.clear();
+
+ vector<long int> positions;
+
+ ifstream inFASTA;
+ openInputFile(filename, inFASTA);
+
+ string input;
+ while(!inFASTA.eof()){
+ input = getline(inFASTA);
+
+ if (input.length() != 0) {
+ if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
+ }
+ }
+ inFASTA.close();
+
+ int numFastaSeqs = positions.size();
+
+ FILE * pFile;
+ long size;
+
+ //get num bytes in file
+ pFile = fopen (filename.c_str(),"rb");
+ if (pFile==NULL) perror ("Error opening file");
+ else{
+ fseek (pFile, 0, SEEK_END);
+ size=ftell (pFile);
+ fclose (pFile);
+ }
+
+ int numSeqsPerProcessor = numFastaSeqs / processors;
+
+ for (int i = 0; i < processors; i++) {
+
+ long int startPos = positions[ i * numSeqsPerProcessor ];
+ if(i == processors - 1){
+ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
+ }else{
+ long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
+ }
+ lines.push_back(new linePair(startPos, numSeqsPerProcessor));
+ }
+
+ return numFastaSeqs;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "setLines");
+ exit(1);
+ }
+}
//***************************************************************************************************************
-void TrimSeqsCommand::getOligos(vector<ofstream*>& outFASTAVec){
+void TrimSeqsCommand::getOligos(vector<string>& outFASTAVec){ //vector<ofstream*>& outFASTAVec
try {
ifstream inOligos;
openInputFile(oligoFile, inOligos);
forPrimer.push_back(oligo);
}
else if(type == "reverse"){
- revPrimer.push_back(oligo);
+ Sequence oligoRC("reverse", oligo);
+ oligoRC.reverseComplement();
+ revPrimer.push_back(oligoRC.getUnaligned());
}
else if(type == "barcode"){
inOligos >> group;
groupVector.push_back(group);
if(allFiles){
- outFASTAVec.push_back(new ofstream((getRootName(fastaFile) + group + ".fasta").c_str(), ios::ate));
+ //outFASTAVec.push_back(new ofstream((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta").c_str(), ios::ate));
+ outputNames.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
+ outFASTAVec.push_back((outputDir + getRootName(getSimpleName(fastaFile)) + group + ".fasta"));
}
}
}
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "getOligos");
+ m->errorOut(e, "TrimSeqsCommand", "getOligos");
exit(1);
}
-
}
-
//***************************************************************************************************************
bool TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
string rawSequence = seq.getUnaligned();
bool success = 0; //guilty until proven innocent
+ //can you find the barcode
for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
string oligo = it->first;
if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
break;
}
}
+
+ //if you found the barcode or if you don't want to allow for diffs
+ if ((bdiffs == 0) || (success == 1)) { return success; }
+
+ else { //try aligning and see if you can find it
+
+ Alignment* alignment;
+ if (barcodes.size() > 0) { //assumes barcodes are all the same length
+ map<string,int>::iterator it=barcodes.begin();
+ string temp = it->first;
+
+ alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (temp.length()+bdiffs+1));
+ }else{ alignment = NULL; }
+
+
+ //can you find the barcode
+ for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
+ string oligo = it->first;
+ int length = oligo.length();
+
+ if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
+ success = 0;
+ break;
+ }
+
+ //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
+ alignment->align(oligo, rawSequence.substr(0,length+bdiffs));
+ oligo = alignment->getSeqAAln();
+ string temp = alignment->getSeqBAln();
+ //cout << "barcode = " << oligo << " raw = " << rawSequence.substr(0,oligo.length()) << " raw aligned = " << temp << endl;
+
+ int newStart=0;
+ if(compareDNASeq(oligo, temp, length, newStart, bdiffs)){
+ group = it->second;
+ seq.setUnaligned(rawSequence.substr(newStart));
+ success = 1;
+ break;
+ }
+ }
+
+ if (alignment != NULL) { delete alignment; }
+ }
return success;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "stripBarcode");
+ m->errorOut(e, "TrimSeqsCommand", "stripBarcode");
exit(1);
}
}
}
+ //if you found the primer or if you don't want to allow for diffs
+ if ((pdiffs == 0) || (success == 1)) { return success; }
+
+ else { //try aligning and see if you can find it
+
+ Alignment* alignment;
+ if (numFPrimers > 0) { alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (forPrimer[0].length()+pdiffs+1)); }
+ else{ alignment = NULL; }
+ //can you find the primer
+ for(int i=0;i<numFPrimers;i++){
+ string oligo = forPrimer[i];
+ int length = oligo.length();
+
+ if(rawSequence.length() < oligo.length()){
+ success = 0;
+ break;
+ }
+
+ //resize if neccessary
+ if ((length+pdiffs+1) > alignment->getnRows()) { alignment->resize(length+pdiffs+1); }
+
+ //use needleman to align first primer.length()+numdiffs of sequence to each primer
+ alignment->align(oligo, rawSequence.substr(0,length+pdiffs));
+ oligo = alignment->getSeqAAln();
+ string temp = alignment->getSeqBAln();
+
+ int newStart = 0;
+ if(compareDNASeq(oligo, temp, length, newStart, pdiffs)){
+ seq.setUnaligned(rawSequence.substr(newStart));
+ success = 1;
+ break;
+ }
+ }
+
+ if (alignment != NULL) { delete alignment; }
+ }
+
return success;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "stripForward");
+ m->errorOut(e, "TrimSeqsCommand", "stripForward");
exit(1);
}
}
}
if(compareDNASeq(oligo, rawSequence.substr(rawSequence.length()-oligo.length(),oligo.length()))){
- seq.setUnaligned(rawSequence.substr(rawSequence.length()-oligo.length()));
+ seq.setUnaligned(rawSequence.substr(0,rawSequence.length()-oligo.length()));
success = 1;
break;
}
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "stripReverse");
+ m->errorOut(e, "TrimSeqsCommand", "stripReverse");
exit(1);
}
}
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "cullLength");
+ m->errorOut(e, "TrimSeqsCommand", "cullLength");
exit(1);
}
return success;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "cullHomoP");
+ m->errorOut(e, "TrimSeqsCommand", "cullHomoP");
exit(1);
}
return success;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "cullAmbigs");
+ m->errorOut(e, "TrimSeqsCommand", "cullAmbigs");
exit(1);
}
for(int i=0;i<length;i++){
if(oligo[i] != seq[i]){
- if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
+ if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C') { success = 0; }
else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { success = 0; }
else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { success = 0; }
else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { success = 0; }
else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { success = 0; }
else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { success = 0; }
- if(success == 0) { break; }
+ if(success == 0) { break; }
}
else{
success = 1;
return success;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "compareDNASeq");
+ m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
exit(1);
}
}
+//***************************************************************************************************************
+
+bool TrimSeqsCommand::compareDNASeq(string oligo, string seq, int numBases, int& end, int diffs){
+ try {
+ bool success = 1;
+ int length = oligo.length();
+ end = numBases;
+ int countBases = 0;
+ int countDiffs = 0;
+
+ if (length != 0) {
+ if ((oligo[0] == '-') || (oligo[0] == '.')) { success = 0; return success; } //no gaps allowed at beginning
+ }
+
+ for(int i=0;i<length;i++){
+
+ if ((oligo[i] != '-') && (oligo[i] != '.')) { countBases++; }
+
+ if(oligo[i] != seq[i]){
+ if(oligo[i] == 'A' || oligo[i] == 'T' || oligo[i] == 'G' || oligo[i] == 'C' || oligo[i] == '-' || oligo[i] == '.') { countDiffs++; }
+ else if((oligo[i] == 'N' || oligo[i] == 'I') && (seq[i] == 'N')) { countDiffs++; }
+ else if(oligo[i] == 'R' && (seq[i] != 'A' && seq[i] != 'G')) { countDiffs++; }
+ else if(oligo[i] == 'Y' && (seq[i] != 'C' && seq[i] != 'T')) { countDiffs++; }
+ else if(oligo[i] == 'M' && (seq[i] != 'C' && seq[i] != 'A')) { countDiffs++; }
+ else if(oligo[i] == 'K' && (seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
+ else if(oligo[i] == 'W' && (seq[i] != 'T' && seq[i] != 'A')) { countDiffs++; }
+ else if(oligo[i] == 'S' && (seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
+ else if(oligo[i] == 'B' && (seq[i] != 'C' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
+ else if(oligo[i] == 'D' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'G')) { countDiffs++; }
+ else if(oligo[i] == 'H' && (seq[i] != 'A' && seq[i] != 'T' && seq[i] != 'C')) { countDiffs++; }
+ else if(oligo[i] == 'V' && (seq[i] != 'A' && seq[i] != 'C' && seq[i] != 'G')) { countDiffs++; }
+
+ if(countDiffs > diffs) { success = 0; break; }
+ }
+ else{
+ success = 1;
+ }
+
+ if (countBases >= numBases) { end = countBases; break; } //stop checking after end of barcode or primer
+ }
+
+ //if it's a success we want to check for total diffs in driver, so save it.
+ if (success == 1) { currentSeqsTdiffs = countDiffs; }
+
+ return success;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
+ exit(1);
+ }
+}
//***************************************************************************************************************
bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){
try {
string rawSequence = seq.getUnaligned();
- int seqLength = rawSequence.length();
- string name;
+ int seqLength; // = rawSequence.length();
+ string name, temp, temp2;
- qFile >> name;
- if (name.length() != 0) { if(name.substr(1) != seq.getName()) { mothurOut("sequence name mismatch btwn fasta and qual file"); mothurOutEndLine(); } }
+ qFile >> name >> temp;
+
+ splitAtEquals(temp2, temp); //separates length=242, temp=length, temp2=242
+ convert(temp, seqLength); //converts string to int
+
+ if (name.length() != 0) { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta and qual file"); m->mothurOutEndLine(); } }
while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
int score;
return 1;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "stripQualThreshold");
+ m->errorOut(e, "TrimSeqsCommand", "stripQualThreshold");
exit(1);
}
}
string name;
qFile >> name;
- if (name[0] == '>') { if(name.substr(1) != seq.getName()) { mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); mothurOutEndLine(); } }
+ if (name[0] == '>') { if(name.substr(1) != seq.getName()) { m->mothurOut("sequence name mismatch btwn fasta: " + seq.getName() + " and qual file: " + name); m->mothurOutEndLine(); } }
while (!qFile.eof()) { char c = qFile.get(); if (c == 10 || c == 13){ break; } }
return success;
}
catch(exception& e) {
- errorOut(e, "TrimSeqsCommand", "cullQualAverage");
+ m->errorOut(e, "TrimSeqsCommand", "cullQualAverage");
exit(1);
}
}
projects / mothur.git / blobdiff