*/
#include "trimseqscommand.h"
+#include "needlemanoverlap.hpp"
//***************************************************************************************************************
else {
//valid paramters for this command
string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
- "qthreshold", "qaverage", "allfiles", "qtrim", "processors", "outputdir","inputdir"};
+ "qthreshold", "qaverage", "allfiles", "qtrim","tdiffs", "pdiffs", "bdiffs", "processors", "outputdir","inputdir"};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
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 = true; }
void TrimSeqsCommand::help(){
try {
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, qtrim and allfiles.\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 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");
if (origSeq != "") {
int group;
string trashCode = "";
+ int currentSeqsDiffs = 0;
if(qFileName != ""){
if(qThreshold != 0) { success = stripQualThreshold(currSeq, qFile); }
if(barcodes.size() != 0){
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'; }
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;
}
}
}
+ //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;
}
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;
}
}
+//***************************************************************************************************************
+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;
+ 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; } }