*
*/
-using namespace std;
-
#include "sequence.hpp"
/***********************************************************************/
-
-Sequence::Sequence() {}
-
+Sequence::Sequence(){
+ m = MothurOut::getInstance();
+ initialize();
+}
/***********************************************************************/
-
Sequence::Sequence(string newName, string sequence) {
- name = newName;
- if(sequence.find_first_of('-') != string::npos) {
+ try {
+ m = MothurOut::getInstance();
+ initialize();
+ name = newName;
+
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
setAligned(sequence);
}
- setUnaligned(sequence);
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
}
+/***********************************************************************/
+Sequence::Sequence(string newName, string sequence, string justUnAligned) {
+ try {
+ m = MothurOut::getInstance();
+ initialize();
+ name = newName;
+
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
+}
+
//********************************************************************************************************************
+//this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
+Sequence::Sequence(istringstream& fastaString){
+ try {
+ m = MothurOut::getInstance();
+
+ initialize();
+ fastaString >> name;
+
+ if (name.length() != 0) {
+
+ name = name.substr(1);
+ string sequence;
+
+ //read comments
+ while ((name[0] == '#') && fastaString) {
+ while (!fastaString.eof()) { char c = fastaString.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+ sequence = getCommentString(fastaString);
+
+ if (fastaString) {
+ fastaString >> name;
+ name = name.substr(1);
+ }else {
+ name = "";
+ break;
+ }
+ }
+
+ while (!fastaString.eof()) { char c = fastaString.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+
+ int numAmbig = 0;
+ sequence = getSequenceString(fastaString, numAmbig);
+
+ setAligned(sequence);
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
+
+ if ((numAmbig / (float) numBases) > 0.25) { m->mothurOut("[WARNING]: We found more than 25% of the bases in sequence " + name + " to be ambiguous. Mothur is not setup to process protein sequences."); m->mothurOutEndLine(); }
+
+ }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaString.tellg()) + ". Blank name."); m->mothurOutEndLine(); }
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
+}
+//********************************************************************************************************************
+//this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
+Sequence::Sequence(istringstream& fastaString, string JustUnaligned){
+ try {
+ m = MothurOut::getInstance();
+
+ initialize();
+ fastaString >> name;
+
+ if (name.length() != 0) {
+
+ name = name.substr(1);
+ string sequence;
+
+ //read comments
+ while ((name[0] == '#') && fastaString) {
+ while (!fastaString.eof()) { char c = fastaString.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+ sequence = getCommentString(fastaString);
+
+ if (fastaString) {
+ fastaString >> name;
+ name = name.substr(1);
+ }else {
+ name = "";
+ break;
+ }
+ }
+
+ while (!fastaString.eof()) { char c = fastaString.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+
+ int numAmbig = 0;
+ sequence = getSequenceString(fastaString, numAmbig);
+
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
+
+ if ((numAmbig / (float) numBases) > 0.25) { m->mothurOut("[WARNING]: We found more than 25% of the bases in sequence " + name + " to be ambiguous. Mothur is not setup to process protein sequences."); m->mothurOutEndLine(); }
+
+ }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaString.tellg()) + ". Blank name."); m->mothurOutEndLine(); }
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
+}
+
+//********************************************************************************************************************
+//this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
Sequence::Sequence(ifstream& fastaFile){
-
- string accession; // provided a file handle to a fasta-formatted sequence file, read in the next
- fastaFile >> accession; // accession number and sequence we find...
- setName(accession);
+ try {
+ m = MothurOut::getInstance();
+ initialize();
+ fastaFile >> name;
+
+ if (name.length() != 0) {
+
+ name = name.substr(1);
+
+ string sequence;
+
+ //read comments
+ while ((name[0] == '#') && fastaFile) {
+ while (!fastaFile.eof()) { char c = fastaFile.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+ sequence = getCommentString(fastaFile);
+
+ if (fastaFile) {
+ fastaFile >> name;
+ name = name.substr(1);
+ }else {
+ name = "";
+ break;
+ }
+ }
+
+ //read real sequence
+ while (!fastaFile.eof()) { char c = fastaFile.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+
+ int numAmbig = 0;
+ sequence = getSequenceString(fastaFile, numAmbig);
+
+ setAligned(sequence);
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
+
+ if ((numAmbig / (float) numBases) > 0.25) { m->mothurOut("[WARNING]: We found more than 25% of the bases in sequence " + name + " to be ambiguous. Mothur is not setup to process protein sequences."); m->mothurOutEndLine(); }
+
+ }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaFile.tellg()) + ". Blank name."); m->mothurOutEndLine(); }
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
+}
+//********************************************************************************************************************
+//this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
+Sequence::Sequence(ifstream& fastaFile, string& extraInfo, bool getInfo){
+ try {
+ m = MothurOut::getInstance();
+ initialize();
+ fastaFile >> name;
+ extraInfo = "";
+
+ if (name.length() != 0) {
+
+ name = name.substr(1);
+
+ string sequence;
+
+ //read comments
+ while ((name[0] == '#') && fastaFile) {
+ while (!fastaFile.eof()) { char c = fastaFile.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+ sequence = getCommentString(fastaFile);
+
+ if (fastaFile) {
+ fastaFile >> name;
+ name = name.substr(1);
+ }else {
+ name = "";
+ break;
+ }
+ }
+
+ //read info after sequence name
+ while (!fastaFile.eof()) {
+ char c = fastaFile.get();
+ if (c == 10 || c == 13){ break; }
+ extraInfo += c;
+ }
+
+ int numAmbig = 0;
+ sequence = getSequenceString(fastaFile, numAmbig);
+
+ setAligned(sequence);
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
+
+ if ((numAmbig / (float) numBases) > 0.25) { m->mothurOut("[WARNING]: We found more than 25% of the bases in sequence " + name + " to be ambiguous. Mothur is not setup to process protein sequences."); m->mothurOutEndLine(); }
+
+ }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaFile.tellg()) + ". Blank name."); m->mothurOutEndLine(); }
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
+}
+//********************************************************************************************************************
+//this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
+Sequence::Sequence(ifstream& fastaFile, string JustUnaligned){
+ try {
+ m = MothurOut::getInstance();
+ initialize();
+ fastaFile >> name;
+
+ if (name.length() != 0) {
+ name = name.substr(1);
+ string sequence;
+
+ //read comments
+ while ((name[0] == '#') && fastaFile) {
+ while (!fastaFile.eof()) { char c = fastaFile.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+ sequence = getCommentString(fastaFile);
+
+ if (fastaFile) {
+ fastaFile >> name;
+ name = name.substr(1);
+ }else {
+ name = "";
+ break;
+ }
+ }
+
+ //read real sequence
+ while (!fastaFile.eof()) { char c = fastaFile.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
+
+ int numAmbig = 0;
+ sequence = getSequenceString(fastaFile, numAmbig);
+
+ //setUnaligned removes any gap characters for us
+ setUnaligned(sequence);
+
+ if ((numAmbig / (float) numBases) > 0.25) { m->mothurOut("[WARNING]: We found more than 25% of the bases in sequence " + name + " to be ambiguous. Mothur is not setup to process protein sequences."); m->mothurOutEndLine(); }
+
+ }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaFile.tellg()) + ". Blank name."); m->mothurOutEndLine(); }
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "Sequence");
+ exit(1);
+ }
+}
- char letter;
- string sequence;
+//********************************************************************************************************************
+string Sequence::getSequenceString(ifstream& fastaFile, int& numAmbig) {
+ try {
+ char letter;
+ string sequence = "";
+ numAmbig = 0;
+
+ while(fastaFile){
+ letter= fastaFile.get();
+ if(letter == '>'){
+ fastaFile.putback(letter);
+ break;
+ }
+ else if(isprint(letter)){
+ letter = toupper(letter);
+ if(letter == 'U'){letter = 'T';}
+ if(letter != '.' && letter != '-' && letter != 'A' && letter != 'T' && letter != 'G' && letter != 'C' && letter != 'N'){
+ letter = 'N';
+ numAmbig++;
+ }
+ sequence += letter;
+ }
+ }
+
+ return sequence;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "getSequenceString");
+ exit(1);
+ }
+}
+//********************************************************************************************************************
+//comment can contain '>' so we need to account for that
+string Sequence::getCommentString(ifstream& fastaFile) {
+ try {
+ char letter;
+ string sequence = "";
+
+ while(fastaFile){
+ letter=fastaFile.get();
+ if((letter == '\r') || (letter == '\n')){
+ m->gobble(fastaFile); //in case its a \r\n situation
+ break;
+ }
+ }
+
+ return sequence;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "getCommentString");
+ exit(1);
+ }
+}
+//********************************************************************************************************************
+string Sequence::getSequenceString(istringstream& fastaFile, int& numAmbig) {
+ try {
+ char letter;
+ string sequence = "";
+ numAmbig = 0;
+
+ while(!fastaFile.eof()){
+ letter= fastaFile.get();
- while(fastaFile){
- letter= fastaFile.get();
- if(letter == '>'){
- fastaFile.putback(letter);
- break;
+ if(letter == '>'){
+ fastaFile.putback(letter);
+ break;
+ }
+ else if(isprint(letter)){
+ letter = toupper(letter);
+ if(letter == 'U'){letter = 'T';}
+ if(letter != '.' && letter != '-' && letter != 'A' && letter != 'T' && letter != 'G' && letter != 'C' && letter != 'N'){
+ letter = 'N';
+ numAmbig++;
+ }
+ sequence += letter;
+ }
}
- else if(isprint(letter)){
- letter = toupper(letter);
- if(letter == 'U'){letter = 'T';}
- sequence += letter;
+
+ return sequence;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "getSequenceString");
+ exit(1);
+ }
+}
+//********************************************************************************************************************
+//comment can contain '>' so we need to account for that
+string Sequence::getCommentString(istringstream& fastaFile) {
+ try {
+ char letter;
+ string sequence = "";
+
+ while(fastaFile){
+ letter=fastaFile.get();
+ if((letter == '\r') || (letter == '\n')){
+ m->gobble(fastaFile); //in case its a \r\n situation
+ break;
+ }
}
+ return sequence;
}
-
- if(sequence.find_first_of('-') != string::npos){ // if there are any gaps in the sequence, assume that it is
- setAligned(sequence); // an alignment file
+ catch(exception& e) {
+ m->errorOut(e, "Sequence", "getCommentString");
+ exit(1);
}
- setUnaligned(sequence); // also set the unaligned sequence file
}
-
//********************************************************************************************************************
-string Sequence::convert2ints() {
+void Sequence::initialize(){
- if(unaligned == "") { /* need to throw an error */ }
+ name = "";
+ unaligned = "";
+ aligned = "";
+ pairwise = "";
- string processed;
+ numBases = 0;
+ alignmentLength = 0;
+ isAligned = 0;
+ startPos = -1;
+ endPos = -1;
+ longHomoPolymer = -1;
+ ambigBases = -1;
- for(int i=0;i<unaligned.length();i++) {
- if(toupper(unaligned[i]) == 'A') { processed += '0'; }
- else if(toupper(unaligned[i]) == 'C') { processed += '1'; }
- else if(toupper(unaligned[i]) == 'G') { processed += '2'; }
- else if(toupper(unaligned[i]) == 'T') { processed += '3'; }
- else if(toupper(unaligned[i]) == 'U') { processed += '3'; }
- else { processed += '4'; }
- }
- return processed;
-}
+}
//********************************************************************************************************************
void Sequence::setUnaligned(string sequence){
- if(sequence.find_first_of('-') != string::npos) {
+ if(sequence.find_first_of('.') != string::npos || sequence.find_first_of('-') != string::npos) {
string temp = "";
for(int j=0;j<sequence.length();j++) {
if(isalpha(sequence[j])) { temp += sequence[j]; }
else {
unaligned = sequence;
}
+ numBases = unaligned.length();
}
//********************************************************************************************************************
void Sequence::setAligned(string sequence){
+
+ //if the alignment starts or ends with a gap, replace it with a period to indicate missing data
aligned = sequence;
+ alignmentLength = aligned.length();
+ setUnaligned(sequence);
+
+ if(aligned[0] == '-'){
+ for(int i=0;i<alignmentLength;i++){
+ if(aligned[i] == '-'){
+ aligned[i] = '.';
+ }
+ else{
+ break;
+ }
+ }
+ for(int i=alignmentLength-1;i>=0;i--){
+ if(aligned[i] == '-'){
+ aligned[i] = '.';
+ }
+ else{
+ break;
+ }
+ }
+ }
+ isAligned = 1;
}
//********************************************************************************************************************
//********************************************************************************************************************
+string Sequence::convert2ints() {
+
+ if(unaligned == "") { /* need to throw an error */ }
+
+ string processed;
+
+ for(int i=0;i<unaligned.length();i++) {
+ if(toupper(unaligned[i]) == 'A') { processed += '0'; }
+ else if(toupper(unaligned[i]) == 'C') { processed += '1'; }
+ else if(toupper(unaligned[i]) == 'G') { processed += '2'; }
+ else if(toupper(unaligned[i]) == 'T') { processed += '3'; }
+ else if(toupper(unaligned[i]) == 'U') { processed += '3'; }
+ else { processed += '4'; }
+ }
+ return processed;
+}
+
+//********************************************************************************************************************
+
string Sequence::getName(){
return name;
}
//********************************************************************************************************************
string Sequence::getAligned(){
- return aligned;
+ if(isAligned == 0) { return unaligned; }
+ else { return aligned; }
}
+//********************************************************************************************************************
+
+string Sequence::getInlineSeq(){
+ return name + '\t' + aligned;
+}
+
+
//********************************************************************************************************************
string Sequence::getPairwise(){
//********************************************************************************************************************
-int Sequence::getLength(){
- if(unaligned.length() > aligned.length())
- return unaligned.length();
- return aligned.length();
+int Sequence::getNumBases(){
+ return numBases;
}
//********************************************************************************************************************
void Sequence::printSequence(ostream& out){
- string toPrint = unaligned;
- if(aligned.length() > unaligned.length())
- toPrint = aligned;
- out << ">" << name << "\n" << toPrint << "\n";
+
+ out << ">" << name << endl;
+ if(isAligned){
+ out << aligned << endl;
+ }
+ else{
+ out << unaligned << endl;
+ }
}
//********************************************************************************************************************
-int Sequence::getUnalignLength(){
- return unaligned.length();
+int Sequence::getAlignLength(){
+ return alignmentLength;
}
//********************************************************************************************************************
-int Sequence::getAlignLength(){
- return aligned.length();
+int Sequence::getAmbigBases(){
+ if(ambigBases == -1){
+ ambigBases = 0;
+ for(int j=0;j<numBases;j++){
+ if(unaligned[j] != 'A' && unaligned[j] != 'T' && unaligned[j] != 'G' && unaligned[j] != 'C'){
+ ambigBases++;
+ }
+ }
+ }
+
+ return ambigBases;
+}
+
+//********************************************************************************************************************
+
+void Sequence::removeAmbigBases(){
+
+ for(int j=0;j<alignmentLength;j++){
+ if(aligned[j] != 'A' && aligned[j] != 'T' && aligned[j] != 'G' && aligned[j] != 'C'){
+ aligned[j] = '-';
+ }
+ }
+ setUnaligned(aligned);
+}
+
+//********************************************************************************************************************
+
+int Sequence::getLongHomoPolymer(){
+ if(longHomoPolymer == -1){
+ longHomoPolymer = 1;
+ int homoPolymer = 1;
+ for(int j=1;j<numBases;j++){
+ if(unaligned[j] == unaligned[j-1]){
+ homoPolymer++;
+ }
+ else{
+ if(homoPolymer > longHomoPolymer){ longHomoPolymer = homoPolymer; }
+ homoPolymer = 1;
+ }
+ }
+ if(homoPolymer > longHomoPolymer){ longHomoPolymer = homoPolymer; }
+ }
+ return longHomoPolymer;
+}
+
+//********************************************************************************************************************
+
+int Sequence::getStartPos(){
+ if(startPos == -1){
+ for(int j = 0; j < alignmentLength; j++) {
+ if(aligned[j] != '.'){
+ startPos = j + 1;
+ break;
+ }
+ }
+ }
+ if(isAligned == 0){ startPos = 1; }
+
+ return startPos;
+}
+
+//********************************************************************************************************************
+
+void Sequence::padToPos(int start){
+
+ for(int j = startPos-1; j < start-1; j++) {
+ aligned[j] = '.';
+ }
+ startPos = start;
+
+}
+//********************************************************************************************************************
+
+int Sequence::filterToPos(int start){
+
+ if (start > aligned.length()) { start = aligned.length(); m->mothurOut("[ERROR]: start to large.\n"); }
+
+ for(int j = 0; j < start-1; j++) {
+ aligned[j] = '.';
+ }
+
+ //things like ......----------AT become ................AT
+ for(int j = start-1; j < aligned.length(); j++) {
+ if (isalpha(aligned[j])) { break; }
+ else { aligned[j] = '.'; }
+ }
+ setUnaligned(aligned);
+
+ return 0;
+
+}
+//********************************************************************************************************************
+
+int Sequence::filterFromPos(int end){
+
+ if (end > aligned.length()) { end = aligned.length(); m->mothurOut("[ERROR]: end to large.\n"); }
+
+ for(int j = end; j < aligned.length(); j++) {
+ aligned[j] = '.';
+ }
+
+ for(int j = aligned.length()-1; j < 0; j--) {
+ if (isalpha(aligned[j])) { break; }
+ else { aligned[j] = '.'; }
+ }
+
+ setUnaligned(aligned);
+
+ return 0;
+}
+//********************************************************************************************************************
+
+int Sequence::getEndPos(){
+ if(endPos == -1){
+ for(int j=alignmentLength-1;j>=0;j--){
+ if(aligned[j] != '.'){
+ endPos = j + 1;
+ break;
+ }
+ }
+ }
+ if(isAligned == 0){ endPos = numBases; }
+
+ return endPos;
}
//********************************************************************************************************************
+void Sequence::padFromPos(int end){
+ cout << end << '\t' << endPos << endl;
+ for(int j = end; j < endPos; j++) {
+ aligned[j] = '.';
+ }
+ endPos = end;
+
+}
+
+//********************************************************************************************************************
+
+bool Sequence::getIsAligned(){
+ return isAligned;
+}
+//********************************************************************************************************************
+
+void Sequence::reverseComplement(){
+
+ string temp;
+ for(int i=numBases-1;i>=0;i--){
+ if(unaligned[i] == 'A') { temp += 'T'; }
+ else if(unaligned[i] == 'T'){ temp += 'A'; }
+ else if(unaligned[i] == 'G'){ temp += 'C'; }
+ else if(unaligned[i] == 'C'){ temp += 'G'; }
+ else { temp += 'N'; }
+ }
+ unaligned = temp;
+ aligned = temp;
+
+}
+//********************************************************************************************************************
+
+void Sequence::trim(int length){
+
+ if(numBases > length){
+ unaligned = unaligned.substr(0,length);
+ numBases = length;
+ }
+
+}
+///**************************************************************************************************/