else {
//valid paramters for this command
string AlignArray[] = {"fasta", "flip", "oligos", "maxambig", "maxhomop", "minlength", "maxlength", "qfile",
- "qthreshold", "qaverage", "allfiles", "qtrim","diffs", "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, "diffs", false); if (temp == "not found") { temp = "0"; }
- convert(temp, diffs);
+
+ 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, "tdiffs", false); if (temp == "not found") { int tempTotal = pdiffs + bdiffs; temp = toString(tempTotal); }
+ convert(temp, tdiffs);
+
+ if(tdiffs == 0){ tdiffs = bdiffs + pdiffs; }
temp = validParameter.validFile(parameters, "qfile", true);
if (temp == "not found") { qFileName = ""; }
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 tdiffs parameter is used to specify the total number of differences allowed in the sequence. The default is pdiffs + bdiffs.\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'; }
+// cout << "here: " << success << endl;
+ if(success > bdiffs){ trashCode += 'b'; }
+ else{ currentSeqsDiffs += success; }
}
-
+
if(numFPrimers != 0){
success = stripForward(currSeq);
- if(!success){ trashCode += 'f'; }
+ if(success > pdiffs){ trashCode += 'f'; }
+ else{ currentSeqsDiffs += success; }
}
-
+
+ if (currentSeqsDiffs > tdiffs) { trashCode += 't'; }
+
if(numRPrimers != 0){
success = stripReverse(currSeq);
if(!success){ trashCode += 'r'; }
}
//***************************************************************************************************************
-bool TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
+int TrimSeqsCommand::stripBarcode(Sequence& seq, int& group){
try {
string rawSequence = seq.getUnaligned();
- bool success = 0; //guilty until proven innocent
+ bool success = bdiffs + 1; //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
- success = 0;
+ success = bdiffs + 1;
break;
}
if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
group = it->second;
seq.setUnaligned(rawSequence.substr(oligo.length()));
- success = 1;
+ success = 0;
break;
}
}
//if you found the barcode or if you don't want to allow for diffs
- if ((diffs == 0) || (success == 1)) { return success; }
+// cout << success;
+ if ((bdiffs == 0) || (success == 0)) { return success; }
else { //try aligning and see if you can find it
-
+// cout << endl;
+
+ int maxLength = 0;
+
Alignment* alignment;
- if (barcodes.size() > 0) { //assumes barcodes are all the same length
+ if (barcodes.size() > 0) {
map<string,int>::iterator it=barcodes.begin();
- string temp = it->first;
-
- alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (temp.length()+diffs+1));
+
+ for(it;it!=barcodes.end();it++){
+ if(it->first.length() > maxLength){
+ maxLength = it->first.length();
+ }
+ }
+ alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (maxLength+bdiffs+1));
+
}else{ alignment = NULL; }
-
//can you find the barcode
+ int minDiff = 1e6;
+ int minCount = 1;
+ int minGroup = -1;
+ int minPos = 0;
+
for(map<string,int>::iterator it=barcodes.begin();it!=barcodes.end();it++){
string oligo = it->first;
- int length = oligo.length();
+// int length = oligo.length();
- if(rawSequence.length() < oligo.length()){ //let's just assume that the barcodes are the same length
- success = 0;
+ if(rawSequence.length() < maxLength){ //let's just assume that the barcodes are the same length
+ success = bdiffs + 1;
break;
}
//use needleman to align first barcode.length()+numdiffs of sequence to each barcode
- alignment->align(oligo, rawSequence.substr(0,length+diffs));
+ alignment->align(oligo, rawSequence.substr(0,oligo.length()+bdiffs));
oligo = alignment->getSeqAAln();
string temp = alignment->getSeqBAln();
- //cout << "barcode = " << oligo << " raw = " << rawSequence.substr(0,oligo.length()) << " raw aligned = " << temp << endl;
+
+ int alnLength = oligo.length();
+
+ for(int i=oligo.length()-1;i>=0;i--){
+ if(oligo[i] != '-'){ alnLength = i+1; break; }
+ }
+ oligo = oligo.substr(0,alnLength);
+ temp = temp.substr(0,alnLength);
+// cout << "barcode = " << oligo << " raw = " << rawSequence.substr(0,alnLength) << " raw aligned = " << temp << endl;
+ cout << seq.getName() << endl;
+ cout << temp << endl;
+ cout << oligo << endl;
+ cout << alnLength << endl;
+ cout << endl;
int newStart=0;
- if(compareDNASeq(oligo, temp, length, newStart)){
- group = it->second;
- seq.setUnaligned(rawSequence.substr(newStart));
- success = 1;
- break;
+ int numDiff = countDiffs(oligo, temp);
+ if(numDiff < minDiff){
+ minDiff = numDiff;
+ minCount = 1;
+ minGroup = it->second;
+ minPos = 0;
+ for(int i=0;i<alnLength;i++){
+ if(temp[i] != '-'){
+ minPos++;
+ }
+ }
+ }
+ else if(numDiff == minDiff){
+ minCount++;
}
+
+ }
+ if(minDiff > bdiffs){ success = bdiffs + 1; }
+ else if(minCount > 1) { success = bdiffs + 1; }
+ else{
+ group = minGroup;
+ seq.setUnaligned("*" + rawSequence.substr(minPos));
+ success = minDiff;
}
if (alignment != NULL) { delete alignment; }
+
}
return success;
//***************************************************************************************************************
-bool TrimSeqsCommand::stripForward(Sequence& seq){
+int TrimSeqsCommand::stripForward(Sequence& seq){
try {
string rawSequence = seq.getUnaligned();
- bool success = 0; //guilty until proven innocent
+ bool success = pdiffs + 1; //guilty until proven innocent
+ //can you find the primer
for(int i=0;i<numFPrimers;i++){
string oligo = forPrimer[i];
-
- if(rawSequence.length() < oligo.length()){
- success = 0;
+
+ if(rawSequence.length() < oligo.length()){
+ success = pdiffs + 1;
break;
}
if(compareDNASeq(oligo, rawSequence.substr(0,oligo.length()))){
seq.setUnaligned(rawSequence.substr(oligo.length()));
- success = 1;
+ success = 0;
break;
}
}
- //if you found the primer or if you don't want to allow for diffs
- if ((diffs == 0) || (success == 1)) { return success; }
+ //if you found the barcode or if you don't want to allow for diffs
+// cout << success;
+ if ((pdiffs == 0) || (success == 0)) { return success; }
else { //try aligning and see if you can find it
-
-
+// cout << endl;
+
+ int maxLength = 0;
+
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; }
+ if (numFPrimers > 0) {
+
+ for(int i=0;i<numFPrimers;i++){
+ if(forPrimer[i].length() > maxLength){
+ maxLength = forPrimer[i].length();
+ }
+ }
+ alignment = new NeedlemanOverlap(-2.0, 1.0, -1.0, (maxLength+pdiffs+1));
+
+ }else{ alignment = NULL; }
+
+ //can you find the barcode
+ int minDiff = 1e6;
+ int minCount = 1;
+ int minPos = 0;
- //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;
+ if(rawSequence.length() < maxLength){
+ success = pdiffs + 1;
break;
}
- //use needleman to align first primer.length()+numdiffs of sequence to each primer
- alignment->align(oligo, rawSequence.substr(0,length+diffs));
+ //use needleman to align first barcode.length()+numdiffs of sequence to each barcode
+ alignment->align(oligo, rawSequence.substr(0,oligo.length()+pdiffs));
oligo = alignment->getSeqAAln();
string temp = alignment->getSeqBAln();
+
+ int alnLength = oligo.length();
- int newStart = 0;
- if(compareDNASeq(oligo, temp, length, newStart)){
- seq.setUnaligned(rawSequence.substr(newStart));
- success = 1;
- break;
+ for(int i=oligo.length()-1;i>=0;i--){
+ if(oligo[i] != '-'){ alnLength = i+1; break; }
+ }
+ oligo = oligo.substr(0,alnLength);
+ temp = temp.substr(0,alnLength);
+// cout << "barcode = " << oligo << " raw = " << rawSequence.substr(0,alnLength) << " raw aligned = " << temp << endl;
+ cout << seq.getName() << endl;
+ cout << temp << endl;
+ cout << oligo << endl;
+ cout << alnLength << endl;
+ cout << endl;
+
+ int newStart=0;
+ int numDiff = countDiffs(oligo, temp);
+ if(numDiff < minDiff){
+ minDiff = numDiff;
+ minCount = 1;
+ minPos = 0;
+ for(int i=0;i<alnLength;i++){
+ if(temp[i] != '-'){
+ minPos++;
+ }
+ }
+ }
+ else if(numDiff == minDiff){
+ minCount++;
}
+
+ }
+ if(minDiff > pdiffs){ success = pdiffs + 1; }
+ else if(minCount > 1) { success = pdiffs + 1; }
+ else{
+ seq.setUnaligned("*" + rawSequence.substr(minPos));
+ success = minDiff;
}
if (alignment != NULL) { delete alignment; }
+
}
-
return success;
-
+
}
catch(exception& e) {
m->errorOut(e, "TrimSeqsCommand", "stripForward");
}
//***************************************************************************************************************
-bool TrimSeqsCommand::compareDNASeq(string oligo, string seq, int numBases, int& end){
+int TrimSeqsCommand::countDiffs(string oligo, string seq){
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] == '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;
+
+ return countDiffs;
}
catch(exception& e) {
- m->errorOut(e, "TrimSeqsCommand", "compareDNASeq");
+ m->errorOut(e, "TrimSeqsCommand", "countDiffs");
exit(1);
}