*/
#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","diffs", "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, "diffs", false); if (temp == "not found") { temp = "0"; }
+ convert(temp, diffs);
+
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 diffs parameter .... 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->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 ((diffs == 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()+diffs+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+diffs));
+ 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, rawSequence.substr(0,oligo.length()), length, newStart)){
+ cout << "found match" << endl;
+ 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 ((diffs == 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()+diffs+1)); } //assumes primers are all the same length
+ 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;
+ }
+
+ //use needleman to align first primer.length()+numdiffs of sequence to each primer
+ alignment->align(oligo, rawSequence.substr(0,length+diffs));
+ oligo = alignment->getSeqAAln();
+
+ int newStart = 0;
+ if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()), length, newStart)){
+ 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){
+ try {
+ bool success = 1;
+ int length = oligo.length();
+ end = length;
+ 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++; }
+
+ //cout << oligo[i] << " " << seq[i] << " diffs = " << countDiffs << " countBases = " << countBases << endl;
+
+ 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 = i; break; } //stop checking after end of barcode or primer
+ }
+
+ return success;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
+ exit(1);
+ }
+
+}
//***************************************************************************************************************
bool TrimSeqsCommand::stripQualThreshold(Sequence& seq, ifstream& qFile){
projects / mothur.git / blobdiff