if (numSeqs == 0) { m->mothurOut("Error reading fasta file...please correct."); m->mothurOutEndLine(); return 0; }
if (diffs > length) { m->mothurOut("Error: diffs is greater than your sequence length."); m->mothurOutEndLine(); return 0; }
- //clear sizes since you only needed this info to build the alignSeqs seqPNode structs
-// sizes.clear();
-
+ string fileroot = outputDir + getRootName(getSimpleName(fastafile));
+ string newFastaFile = fileroot + "precluster" + getExtension(fastafile);
+ string newNamesFile = fileroot + "precluster.names";
+ ofstream outFasta;
+ ofstream outNames;
+
+ openOutputFile(newFastaFile, outFasta);
+ openOutputFile(newNamesFile, outNames);
+
//sort seqs by number of identical seqs
- sort(alignSeqs.begin(), alignSeqs.end(), comparePriority);
+ alignSeqs.sort(comparePriority);
int count = 0;
+ int i = 0;
//think about running through twice...
- for (int i = 0; i < numSeqs; i++) {
-
- //are you active
- // itActive = active.find(alignSeqs[i].seq.getName());
-
- if (alignSeqs[i].active) { //this sequence has not been merged yet
-
- //try to merge it with all smaller seqs
- for (int j = i+1; j < numSeqs; j++) {
+ list<seqPNode>::iterator itList;
+ list<seqPNode>::iterator itList2;
+ for (itList = alignSeqs.begin(); itList != alignSeqs.end(); itList++) {
+ //try to merge it with all smaller seqs
+ for (itList2 = alignSeqs.begin(); itList2 != alignSeqs.end(); itList2++) {
- if (m->control_pressed) { return 0; }
+ if (m->control_pressed) { outFasta.close(); outNames.close(); remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; }
- if (alignSeqs[j].active) { //this sequence has not been merged yet
- //are you within "diff" bases
- int mismatch = calcMisMatches(alignSeqs[i].seq.getAligned(), alignSeqs[j].seq.getAligned());
+ if ((*itList).seq.getName() != (*itList2).seq.getName()) { //you don't want to merge with yourself
+ //are you within "diff" bases
+ int mismatch = calcMisMatches((*itList).seq.getAligned(), (*itList2).seq.getAligned());
- if (mismatch <= diffs) {
- //merge
- alignSeqs[i].names += ',' + alignSeqs[j].names;
- alignSeqs[i].numIdentical += alignSeqs[j].numIdentical;
+ if (mismatch <= diffs) {
+ //merge
+ (*itList).names += ',' + (*itList2).names;
+ (*itList).numIdentical += (*itList2).numIdentical;
- alignSeqs[j].active = 0;
- alignSeqs[j].numIdentical = 0;
- count++;
- }
- }//end if j active
- }//end if i != j
+ alignSeqs.erase(itList2); itList2--;
+ count++;
+ }
+ }
+ }
+
+ //ouptut this sequence
+ printData(outFasta, outNames, (*itList));
+
+ //remove sequence
+ alignSeqs.erase(itList); itList--;
- //remove from active list
- alignSeqs[i].active = 0;
- }//end if active i
- if(i % 100 == 0) { cout << i << '\t' << numSeqs - count << '\t' << count << endl; }
+ i++;
+
+ if(i % 100 == 0) { m->mothurOut(toString(i) + "\t" + toString(numSeqs - count) + "\t" + toString(count)); m->mothurOutEndLine(); }
}
+
+ if(i % 100 != 0) { m->mothurOut(toString(i) + "\t" + toString(numSeqs - count) + "\t" + toString(count)); m->mothurOutEndLine(); }
- string fileroot = outputDir + getRootName(getSimpleName(fastafile));
-
- string newFastaFile = fileroot + "precluster" + getExtension(fastafile);
- string newNamesFile = fileroot + "precluster.names";
+ outFasta.close();
+ outNames.close();
- if (m->control_pressed) { return 0; }
+ if (m->control_pressed) { remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; }
+
- m->mothurOut("Total number of sequences before precluster was " + toString(alignSeqs.size()) + "."); m->mothurOutEndLine();
+ m->mothurOut("Total number of sequences before precluster was " + toString(numSeqs) + "."); m->mothurOutEndLine();
m->mothurOut("pre.cluster removed " + toString(count) + " sequences."); m->mothurOutEndLine();
- printData(newFastaFile, newNamesFile);
if (m->control_pressed) { remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; }
else{
seqPNode tempNode(itSize->second, seq, names[seq.getName()]);
alignSeqs.push_back(tempNode);
- if (seq.getAligned().length() > length) { length = alignSeqs[0].seq.getAligned().length(); }
+ if (seq.getAligned().length() > length) { length = seq.getAligned().length(); }
}
}else { //no names file, you are identical to yourself
seqPNode tempNode(1, seq, seq.getName());
alignSeqs.push_back(tempNode);
- if (seq.getAligned().length() > length) { length = alignSeqs[0].seq.getAligned().length(); }
+ if (seq.getAligned().length() > length) { length = seq.getAligned().length(); }
}
}
}
/**************************************************************************************************/
-void PreClusterCommand::printData(string newfasta, string newname){
+void PreClusterCommand::printData(ofstream& outFasta, ofstream& outNames, seqPNode thisSeq){
try {
- ofstream outFasta;
- ofstream outNames;
-
- openOutputFile(newfasta, outFasta);
- openOutputFile(newname, outNames);
-
-
- for (int i = 0; i < alignSeqs.size(); i++) {
- if (alignSeqs[i].numIdentical != 0) {
- alignSeqs[i].seq.printSequence(outFasta);
- outNames << alignSeqs[i].seq.getName() << '\t' << alignSeqs[i].names << endl;
- }
- }
-
- outFasta.close();
- outNames.close();
-
+ thisSeq.seq.printSequence(outFasta);
+ outNames << thisSeq.seq.getName() << '\t' << thisSeq.names << endl;
}
catch(exception& e) {
m->errorOut(e, "PreClusterCommand", "printData");
exit(1);
}
}
-
/**************************************************************************************************/
void PreClusterCommand::readNameFile(){