//**********************************************************************************************************************
ScreenSeqsCommand::ScreenSeqsCommand(){
try {
- abort = true;
- //initialize outputTypes
+ abort = true; calledHelp = true;
vector<string> tempOutNames;
outputTypes["fasta"] = tempOutNames;
outputTypes["name"] = tempOutNames;
ScreenSeqsCommand::ScreenSeqsCommand(string option) {
try {
- abort = false;
+ abort = false; calledHelp = false;
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true; }
else {
//valid paramters for this command
else if (groupfile == "not found") { groupfile = ""; }
namefile = validParameter.validFile(parameters, "name", true);
- if (namefile == "not open") { abort = true; }
+ if (namefile == "not open") { namefile = ""; abort = true; }
else if (namefile == "not found") { namefile = ""; }
alignreport = validParameter.validFile(parameters, "alignreport", true);
int ScreenSeqsCommand::execute(){
try{
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
- //if the user want to optimize we need to no the 90% mark
+ //if the user want to optimize we need to know the 90% mark
vector<unsigned long int> positions;
- if (optimize.size() != 0) { getSummary(positions); } //get summary is paralellized so we need to divideFile, no need to do this step twice so I moved it here
+ if (optimize.size() != 0) { //get summary is paralellized so we need to divideFile, no need to do this step twice so I moved it here
+ //use the namefile to optimize correctly
+ if (namefile != "") { nameMap = m->readNames(namefile); }
+ getSummary(positions);
+ }
else {
positions = m->divideFile(fastafile, processors);
for (int i = 0; i < (positions.size()-1); i++) {
int start = time(NULL);
#ifdef USE_MPI
- int pid, end, numSeqsPerProcessor;
+ int pid, numSeqsPerProcessor;
int tag = 2001;
vector<unsigned long int> MPIPos;
outputNames.push_back(goodNameFile); outputTypes["name"].push_back(goodNameFile);
ofstream goodNameOut; m->openOutputFile(goodNameFile, goodNameOut);
-
+
while(!inputNames.eof()){
if (m->control_pressed) { goodNameOut.close(); inputNames.close(); remove(goodNameFile.c_str()); return 0; }
inputNames >> seqName >> seqList;
it = badSeqNames.find(seqName);
-
+
if(it != badSeqNames.end()){
badSeqNames.erase(it);
if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); remove(goodNameFile.c_str()); remove(goodGroupFile.c_str()); return 0; }
inputGroups >> seqName >> group;
-
+
it = badSeqGroups.find(seqName);
if(it != badSeqGroups.end()){
}
}
}
-
+
+
return 0;
}
sort(ambigBases.begin(), ambigBases.end());
sort(longHomoPolymer.begin(), longHomoPolymer.end());
- int criteriaPercentile = int(numSeqs * (criteria / (float) 100));
+ //numSeqs is the number of unique seqs, startPosition.size() is the total number of seqs, we want to optimize using all seqs
+ int criteriaPercentile = int(startPosition.size() * (criteria / (float) 100));
for (int i = 0; i < optimize.size(); i++) {
if (optimize[i] == "start") { startPos = startPosition[criteriaPercentile]; m->mothurOut("Optimizing start to " + toString(startPos) + "."); m->mothurOutEndLine(); }
- else if (optimize[i] == "end") { int endcriteriaPercentile = int(numSeqs * ((100 - criteria) / (float) 100)); endPos = endPosition[endcriteriaPercentile]; m->mothurOut("Optimizing end to " + toString(endPos) + "."); m->mothurOutEndLine();}
+ else if (optimize[i] == "end") { int endcriteriaPercentile = int(endPosition.size() * ((100 - criteria) / (float) 100)); endPos = endPosition[endcriteriaPercentile]; m->mothurOut("Optimizing end to " + toString(endPos) + "."); m->mothurOutEndLine();}
else if (optimize[i] == "maxambig") { maxAmbig = ambigBases[criteriaPercentile]; m->mothurOut("Optimizing maxambig to " + toString(maxAmbig) + "."); m->mothurOutEndLine(); }
else if (optimize[i] == "maxhomop") { maxHomoP = longHomoPolymer[criteriaPercentile]; m->mothurOut("Optimizing maxhomop to " + toString(maxHomoP) + "."); m->mothurOutEndLine(); }
- else if (optimize[i] == "minlength") { int mincriteriaPercentile = int(numSeqs * ((100 - criteria) / (float) 100)); minLength = seqLength[mincriteriaPercentile]; m->mothurOut("Optimizing minlength to " + toString(minLength) + "."); m->mothurOutEndLine(); }
+ else if (optimize[i] == "minlength") { int mincriteriaPercentile = int(seqLength.size() * ((100 - criteria) / (float) 100)); minLength = seqLength[mincriteriaPercentile]; m->mothurOut("Optimizing minlength to " + toString(minLength) + "."); m->mothurOutEndLine(); }
else if (optimize[i] == "maxlength") { maxLength = seqLength[criteriaPercentile]; m->mothurOut("Optimizing maxlength to " + toString(maxLength) + "."); m->mothurOutEndLine(); }
}
Sequence current(in); m->gobble(in);
if (current.getName() != "") {
- startPosition.push_back(current.getStartPos());
- endPosition.push_back(current.getEndPos());
- seqLength.push_back(current.getNumBases());
- ambigBases.push_back(current.getAmbigBases());
- longHomoPolymer.push_back(current.getLongHomoPolymer());
+ int num = 1;
+ if (namefile != "") {
+ //make sure this sequence is in the namefile, else error
+ map<string, int>::iterator it = nameMap.find(current.getName());
+
+ if (it == nameMap.end()) { m->mothurOut("[ERROR]: " + current.getName() + " is not in your namefile, please correct."); m->mothurOutEndLine(); m->control_pressed = true; }
+ else { num = it->second; }
+ }
+
+ //for each sequence this sequence represents
+ for (int i = 0; i < num; i++) {
+ startPosition.push_back(current.getStartPos());
+ endPosition.push_back(current.getEndPos());
+ seqLength.push_back(current.getNumBases());
+ ambigBases.push_back(current.getAmbigBases());
+ longHomoPolymer.push_back(current.getLongHomoPolymer());
+ }
+
count++;
}
m->openOutputFile(tempFile, out);
out << num << endl;
+ out << startPosition.size() << endl;
for (int k = 0; k < startPosition.size(); k++) { out << startPosition[k] << '\t'; } out << endl;
for (int k = 0; k < endPosition.size(); k++) { out << endPosition[k] << '\t'; } out << endl;
for (int k = 0; k < seqLength.size(); k++) { out << seqLength[k] << '\t'; } out << endl;
int temp, tempNum;
in >> tempNum; m->gobble(in); num += tempNum;
+ in >> tempNum; m->gobble(in);
for (int k = 0; k < tempNum; k++) { in >> temp; startPosition.push_back(temp); } m->gobble(in);
for (int k = 0; k < tempNum; k++) { in >> temp; endPosition.push_back(temp); } m->gobble(in);
for (int k = 0; k < tempNum; k++) { in >> temp; seqLength.push_back(temp); } m->gobble(in);
projects / mothur.git / blobdiff