X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=sequence.cpp;h=d6073d75da3ed630b27b626bbba8615d79b58858;hp=423b9058b23b0815326465f86b9916777c0359ce;hb=a8e2df1b96a57f5f29576b08361b86a96a8eff4f;hpb=5b7ac70116137b52dd7884b76c5bca660a5fea38 diff --git a/sequence.cpp b/sequence.cpp index 423b905..d6073d7 100644 --- a/sequence.cpp +++ b/sequence.cpp @@ -7,60 +7,430 @@ * */ -using namespace std; - #include "sequence.hpp" /***********************************************************************/ - Sequence::Sequence(){ + m = MothurOut::getInstance(); initialize(); } - /***********************************************************************/ - Sequence::Sequence(string newName, string sequence) { - - initialize(); - name = newName; - if(sequence.find_first_of('-') != string::npos) { + try { + m = MothurOut::getInstance(); + initialize(); + name = newName; + + m->checkName(name); + + //setUnaligned removes any gap characters for us + setUnaligned(sequence); setAligned(sequence); - isAligned = 1; } - 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; + + m->checkName(name); + + //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(); + name = getSequenceName(fastaString); + + if (!m->control_pressed) { + 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(); } + } + + } + 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(); + name = getSequenceName(fastaString); + + if (!m->control_pressed) { + 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(); } + + } + + } + 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(); + name = getSequenceName(fastaFile); + + if (!m->control_pressed) { + + 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(); } + + } - char letter; - string 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(ifstream& fastaFile, string& extraInfo, bool getInfo){ + try { + m = MothurOut::getInstance(); + initialize(); + extraInfo = ""; + + name = getSequenceName(fastaFile); + + if (!m->control_pressed) { + 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 || c == -1){ 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(); } + } + + } + 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(); + name = getSequenceName(fastaFile); + + if (!m->control_pressed) { + 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(); } + + } + + } + catch(exception& e) { + m->errorOut(e, "Sequence", "Sequence"); + exit(1); + } +} +//******************************************************************************************************************** +string Sequence::getSequenceName(ifstream& fastaFile) { + try { + string name = ""; + + fastaFile >> name; + + if (name.length() != 0) { + + name = name.substr(1); + + m->checkName(name); + + }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaFile.tellg()) + ". Blank name."); m->mothurOutEndLine(); m->control_pressed = true; } + + return name; + } + catch(exception& e) { + m->errorOut(e, "Sequence", "getSequenceName"); + exit(1); + } +} +//******************************************************************************************************************** +string Sequence::getSequenceName(istringstream& fastaFile) { + try { + string name = ""; + + fastaFile >> name; + + if (name.length() != 0) { + + name = name.substr(1); + + m->checkName(name); + + }else{ m->mothurOut("Error in reading your fastafile, at position " + toString(fastaFile.tellg()) + ". Blank name."); m->mothurOutEndLine(); m->control_pressed = true; } + + return name; + } + catch(exception& e) { + m->errorOut(e, "Sequence", "getSequenceName"); + exit(1); + } +} +//******************************************************************************************************************** +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 (letter == ' ') {;} + 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 (letter == ' ') {;} + 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 } - //******************************************************************************************************************** void Sequence::initialize(){ @@ -75,8 +445,8 @@ void Sequence::initialize(){ isAligned = 0; startPos = -1; endPos = -1; - longHomoPolymer = 0; - ambigBases = 0; + longHomoPolymer = -1; + ambigBases = -1; } @@ -91,7 +461,7 @@ void Sequence::setName(string seqName) { 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 aligned.length()) { start = aligned.length(); m->mothurOut("[ERROR]: start to large.\n"); } + + for(int j = 0; j < start; j++) { + aligned[j] = '.'; + } + + //things like ......----------AT become ................AT + for(int j = start; 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; + if((aligned[j] != '.')&&(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; + aligned = ""; + isAligned = 0; + } + +} + +///**************************************************************************************************/