int ClusterDoturCommand::execute(){
try {
if (abort) { if (calledHelp) { return 0; } return 2; }
ClusterClassic* cluster = new ClusterClassic(cutoff, method, sim);
NameAssignment* nameMap = NULL;
CountTable* ct = NULL;
map<string, int> counts;
if(namefile != "") {
nameMap = new NameAssignment(namefile);
nameMap->readMap();
cluster->readPhylipFile(phylipfile, nameMap);
delete nameMap;
}else if (countfile != "") {
ct = new CountTable();
ct->readTable(countfile, false, false);
cluster->readPhylipFile(phylipfile, ct);
counts = ct->getNameMap();
delete ct;
}else {
cluster->readPhylipFile(phylipfile, nameMap);
}
tag = cluster->getTag();
if (m->getControl_pressed()) { delete cluster; return 0; }
list = cluster->getListVector();
rabund = cluster->getRAbundVector();
if (outputDir == "") { outputDir += util.hasPath(phylipfile); }
fileroot = outputDir + util.getRootName(util.getSimpleName(phylipfile));
map<string, string> variables;
variables["[filename]"] = fileroot;
variables["[clustertag]"] = tag;
string sabundFileName = getOutputFileName("sabund", variables);
string rabundFileName = getOutputFileName("rabund", variables);
//if (countfile != "") { variables["[tag2]"] = "unique_list"; }
string listFileName = getOutputFileName("list", variables);
if (countfile == "") {
util.openOutputFile(sabundFileName, sabundFile);
util.openOutputFile(rabundFileName, rabundFile);
outputNames.push_back(sabundFileName); outputTypes["sabund"].push_back(sabundFileName);
outputNames.push_back(rabundFileName); outputTypes["rabund"].push_back(rabundFileName);
}
util.openOutputFile(listFileName, listFile);
outputNames.push_back(listFileName); outputTypes["list"].push_back(listFileName);
float previousDist = 0.00000;
float rndPreviousDist = 0.00000;
oldRAbund = *rabund;
oldList = *list;
bool printHeaders = true;
int estart = time(NULL); int loop = 0;
while ((cluster->getSmallDist() <= cutoff) && (cluster->getNSeqs() > 1)){
if (m->getControl_pressed()) { delete cluster; delete list; delete rabund; if(countfile == "") {rabundFile.close(); sabundFile.close(); util.mothurRemove((fileroot+ tag + ".rabund")); util.mothurRemove((fileroot+ tag + ".sabund")); }
listFile.close(); util.mothurRemove((fileroot+ tag + ".list")); outputTypes.clear(); return 0; }
cluster->update(cutoff);
float dist = cluster->getSmallDist();
float rndDist = util.ceilDist(dist, precision);
//cout << loop << '\t' << dist << '\t' << oldList.getNumBins() << endl; loop++;
if(previousDist <= 0.0000 && dist != previousDist) { printData("unique", counts, printHeaders); }
else if(rndDist != rndPreviousDist) { printData(toString(rndPreviousDist, length-1), counts, printHeaders); }
previousDist = dist;
rndPreviousDist = rndDist;
oldRAbund = *rabund;
oldList = *list;
}
if(previousDist <= 0.0000) { printData("unique", counts, printHeaders); }
else if(rndPreviousDist<cutoff) { printData(toString(rndPreviousDist, length-1), counts, printHeaders); }
if (countfile == "") {
sabundFile.close();
rabundFile.close();
}
listFile.close();
delete cluster; delete list; delete rabund;
//set list file as new current listfile
string currentName = "";
itTypes = outputTypes.find("list");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setListFile(currentName); } }
//set rabund file as new current rabundfile
itTypes = outputTypes.find("rabund");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setRabundFile(currentName); } }
//set sabund file as new current sabundfile
itTypes = outputTypes.find("sabund");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setSabundFile(currentName); } }
m->mothurOut("\nOutput File Names: \n");
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i] +"\n"); } m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
m->errorOut(e, "ClusterDoturCommand", "execute");
exit(1);
}
}
//***************************************************************************************************************
int SummaryQualCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
int start = time(NULL);
int numSeqs = 0;
vector<int> position;
vector<int> averageQ;
vector< vector<int> > scores;
if (m->control_pressed) { return 0; }
if (namefile != "") { nameMap = m->readNames(namefile); }
else if (countfile != "") {
CountTable ct;
ct.readTable(countfile, false, false);
nameMap = ct.getNameMap();
}
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(qualfile, processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(linePair(positions[i], positions[(i+1)])); }
#else
if (processors == 1) {
lines.push_back(linePair(0, 1000));
}else {
positions = m->setFilePosFasta(qualfile, numSeqs);
if (numSeqs < processors) { processors = numSeqs; }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
lines.push_back(linePair(positions[startIndex], numSeqsPerProcessor));
}
}
#endif
if(processors == 1){ numSeqs = driverCreateSummary(position, averageQ, scores, qualfile, lines[0]); }
else{ numSeqs = createProcessesCreateSummary(position, averageQ, scores, qualfile); }
if (m->control_pressed) { return 0; }
//print summary file
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(qualfile));
string summaryFile = getOutputFileName("summary",variables);
printQual(summaryFile, position, averageQ, scores);
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
//output results to screen
cout.setf(ios::fixed, ios::floatfield); cout.setf(ios::showpoint);
m->mothurOutEndLine();
m->mothurOut("Position\tNumSeqs\tAverageQ"); m->mothurOutEndLine();
for (int i = 0; i < position.size(); i+=100) {
float average = averageQ[i] / (float) position[i];
cout << i << '\t' << position[i] << '\t' << average;
m->mothurOutJustToLog(toString(i) + "\t" + toString(position[i]) + "\t" + toString(average)); m->mothurOutEndLine();
}
m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile);
m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
m->errorOut(e, "SummaryQualCommand", "execute");
exit(1);
}
}
int SeqSummaryCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
//set current fasta to fastafile
m->setFastaFile(fastafile);
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile));
string summaryFile = getOutputFileName("summary",variables);
int numSeqs = 0;
vector<int> startPosition;
vector<int> endPosition;
vector<int> seqLength;
vector<int> ambigBases;
vector<int> longHomoPolymer;
if (namefile != "") { nameMap = m->readNames(namefile); }
else if (countfile != "") {
CountTable ct;
ct.readTable(countfile, false, false);
nameMap = ct.getNameMap();
}
if (m->control_pressed) { return 0; }
#ifdef USE_MPI
int pid, numSeqsPerProcessor;
int tag = 2001;
int startTag = 1; int endTag = 2; int lengthTag = 3; int baseTag = 4; int lhomoTag = 5;
int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
vector<unsigned long long> MPIPos;
MPI_Status status;
MPI_Status statusOut;
MPI_File inMPI;
MPI_File outMPI;
MPI_Comm_size(MPI_COMM_WORLD, &processors);
MPI_Comm_rank(MPI_COMM_WORLD, &pid);
char tempFileName[1024];
strcpy(tempFileName, fastafile.c_str());
char sumFileName[1024];
strcpy(sumFileName, summaryFile.c_str());
MPI_File_open(MPI_COMM_WORLD, tempFileName, MPI_MODE_RDONLY, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
MPI_File_open(MPI_COMM_WORLD, sumFileName, outMode, MPI_INFO_NULL, &outMPI);
if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); return 0; }
if (pid == 0) { //you are the root process
//print header
string outputString = "seqname\tstart\tend\tnbases\tambigs\tpolymer\tnumSeqs\n";
int length = outputString.length();
char* buf2 = new char[length];
memcpy(buf2, outputString.c_str(), length);
MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &statusOut);
delete buf2;
MPIPos = m->setFilePosFasta(fastafile, numSeqs); //fills MPIPos, returns numSeqs
for(int i = 1; i < processors; i++) {
MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
}
//figure out how many sequences you have to do
numSeqsPerProcessor = numSeqs / processors;
int startIndex = pid * numSeqsPerProcessor;
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
//do your part
MPICreateSummary(startIndex, numSeqsPerProcessor, startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, inMPI, outMPI, MPIPos);
}else { //i am the child process
MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPIPos.resize(numSeqs+1);
MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
//figure out how many sequences you have to align
numSeqsPerProcessor = numSeqs / processors;
int startIndex = pid * numSeqsPerProcessor;
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
//do your part
MPICreateSummary(startIndex, numSeqsPerProcessor, startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, inMPI, outMPI, MPIPos);
}
MPI_File_close(&inMPI);
MPI_File_close(&outMPI);
MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
if (pid == 0) {
//get the info from the child processes
for(int i = 1; i < processors; i++) {
int size;
MPI_Recv(&size, 1, MPI_INT, i, tag, MPI_COMM_WORLD, &status);
vector<int> temp; temp.resize(size+1);
for(int j = 0; j < 5; j++) {
MPI_Recv(&temp[0], (size+1), MPI_INT, i, 2001, MPI_COMM_WORLD, &status);
int receiveTag = temp[temp.size()-1]; //child process added a int to the end to indicate what count this is for
if (receiveTag == startTag) {
for (int k = 0; k < size; k++) { startPosition.push_back(temp[k]); }
}else if (receiveTag == endTag) {
for (int k = 0; k < size; k++) { endPosition.push_back(temp[k]); }
}else if (receiveTag == lengthTag) {
for (int k = 0; k < size; k++) { seqLength.push_back(temp[k]); }
}else if (receiveTag == baseTag) {
for (int k = 0; k < size; k++) { ambigBases.push_back(temp[k]); }
}else if (receiveTag == lhomoTag) {
for (int k = 0; k < size; k++) { longHomoPolymer.push_back(temp[k]); }
}
}
}
}else{
//send my counts
int size = startPosition.size();
MPI_Send(&size, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
startPosition.push_back(startTag);
int ierr = MPI_Send(&(startPosition[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
endPosition.push_back(endTag);
ierr = MPI_Send (&(endPosition[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
seqLength.push_back(lengthTag);
ierr = MPI_Send(&(seqLength[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
ambigBases.push_back(baseTag);
ierr = MPI_Send(&(ambigBases[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
longHomoPolymer.push_back(lhomoTag);
ierr = MPI_Send(&(longHomoPolymer[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
}
MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
#else
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(fastafile, processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); }
#else
positions = m->setFilePosFasta(fastafile, numSeqs);
if (positions.size() < processors) { processors = positions.size(); }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor));
}
#endif
if(processors == 1){
numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile, lines[0]);
}else{
numSeqs = createProcessesCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile);
}
if (m->control_pressed) { return 0; }
#endif
#ifdef USE_MPI
if (pid == 0) {
#endif
sort(startPosition.begin(), startPosition.end());
sort(endPosition.begin(), endPosition.end());
sort(seqLength.begin(), seqLength.end());
sort(ambigBases.begin(), ambigBases.end());
sort(longHomoPolymer.begin(), longHomoPolymer.end());
int size = startPosition.size();
//find means
unsigned long long meanStartPosition, meanEndPosition, meanSeqLength, meanAmbigBases, meanLongHomoPolymer;
meanStartPosition = 0; meanEndPosition = 0; meanSeqLength = 0; meanAmbigBases = 0; meanLongHomoPolymer = 0;
for (int i = 0; i < size; i++) {
meanStartPosition += startPosition[i];
meanEndPosition += endPosition[i];
meanSeqLength += seqLength[i];
meanAmbigBases += ambigBases[i];
meanLongHomoPolymer += longHomoPolymer[i];
}
double meanstartPosition, meanendPosition, meanseqLength, meanambigBases, meanlongHomoPolymer;
meanstartPosition = meanStartPosition / (double) size; meanendPosition = meanEndPosition /(double) size; meanlongHomoPolymer = meanLongHomoPolymer / (double) size; meanseqLength = meanSeqLength / (double) size; meanambigBases = meanAmbigBases /(double) size;
int ptile0_25 = int(size * 0.025);
int ptile25 = int(size * 0.250);
int ptile50 = int(size * 0.500);
int ptile75 = int(size * 0.750);
int ptile97_5 = int(size * 0.975);
int ptile100 = size - 1;
//to compensate for blank sequences that would result in startPosition and endPostion equalling -1
if (startPosition[0] == -1) { startPosition[0] = 0; }
if (endPosition[0] == -1) { endPosition[0] = 0; }
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("\t\tStart\tEnd\tNBases\tAmbigs\tPolymer\tNumSeqs"); m->mothurOutEndLine();
m->mothurOut("Minimum:\t" + toString(startPosition[0]) + "\t" + toString(endPosition[0]) + "\t" + toString(seqLength[0]) + "\t" + toString(ambigBases[0]) + "\t" + toString(longHomoPolymer[0]) + "\t" + toString(1)); m->mothurOutEndLine();
m->mothurOut("2.5%-tile:\t" + toString(startPosition[ptile0_25]) + "\t" + toString(endPosition[ptile0_25]) + "\t" + toString(seqLength[ptile0_25]) + "\t" + toString(ambigBases[ptile0_25]) + "\t"+ toString(longHomoPolymer[ptile0_25]) + "\t" + toString(ptile0_25+1)); m->mothurOutEndLine();
m->mothurOut("25%-tile:\t" + toString(startPosition[ptile25]) + "\t" + toString(endPosition[ptile25]) + "\t" + toString(seqLength[ptile25]) + "\t" + toString(ambigBases[ptile25]) + "\t" + toString(longHomoPolymer[ptile25]) + "\t" + toString(ptile25+1)); m->mothurOutEndLine();
m->mothurOut("Median: \t" + toString(startPosition[ptile50]) + "\t" + toString(endPosition[ptile50]) + "\t" + toString(seqLength[ptile50]) + "\t" + toString(ambigBases[ptile50]) + "\t" + toString(longHomoPolymer[ptile50]) + "\t" + toString(ptile50+1)); m->mothurOutEndLine();
m->mothurOut("75%-tile:\t" + toString(startPosition[ptile75]) + "\t" + toString(endPosition[ptile75]) + "\t" + toString(seqLength[ptile75]) + "\t" + toString(ambigBases[ptile75]) + "\t" + toString(longHomoPolymer[ptile75]) + "\t" + toString(ptile75+1)); m->mothurOutEndLine();
m->mothurOut("97.5%-tile:\t" + toString(startPosition[ptile97_5]) + "\t" + toString(endPosition[ptile97_5]) + "\t" + toString(seqLength[ptile97_5]) + "\t" + toString(ambigBases[ptile97_5]) + "\t" + toString(longHomoPolymer[ptile97_5]) + "\t" + toString(ptile97_5+1)); m->mothurOutEndLine();
m->mothurOut("Maximum:\t" + toString(startPosition[ptile100]) + "\t" + toString(endPosition[ptile100]) + "\t" + toString(seqLength[ptile100]) + "\t" + toString(ambigBases[ptile100]) + "\t" + toString(longHomoPolymer[ptile100]) + "\t" + toString(ptile100+1)); m->mothurOutEndLine();
m->mothurOut("Mean:\t" + toString(meanstartPosition) + "\t" + toString(meanendPosition) + "\t" + toString(meanseqLength) + "\t" + toString(meanambigBases) + "\t" + toString(meanlongHomoPolymer)); m->mothurOutEndLine();
if ((namefile == "") && (countfile == "")) { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); }
else { m->mothurOut("# of unique seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); m->mothurOut("total # of seqs:\t" + toString(startPosition.size())); m->mothurOutEndLine(); }
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile);
m->mothurOutEndLine();
#ifdef USE_MPI
}
#endif
//set fasta file as new current fastafile
string current = "";
itTypes = outputTypes.find("summary");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSummaryFile(current); }
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "SeqSummaryCommand", "execute");
exit(1);
}
}
//**********************************************************************************************************************
SharedCommand::SharedCommand(string option) {
try {
abort = false; calledHelp = false; pickedGroups=false;
allLines = 1;
//allow user to run help
if(option == "help") { help(); abort = true; calledHelp = true; }
else if(option == "citation") { citation(); abort = true; calledHelp = true;}
else {
vector<string> myArray = setParameters();
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
ValidParameters validParameter;
map<string, string>::iterator it;
//check to make sure all parameters are valid for command
for (it = parameters.begin(); it != parameters.end(); it++) {
if (!validParameter.isValidParameter(it->first, myArray, it->second)) { abort = true; }
}
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.valid(parameters, "inputdir");
if (inputDir == "not found"){ inputDir = ""; }
else {
string path;
it = parameters.find("list");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["list"] = inputDir + it->second; }
}
it = parameters.find("group");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["group"] = inputDir + it->second; }
}
it = parameters.find("count");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["count"] = inputDir + it->second; }
}
it = parameters.find("biom");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["biom"] = inputDir + it->second; }
}
}
vector<string> tempOutNames;
outputTypes["shared"] = tempOutNames;
outputTypes["group"] = tempOutNames;
outputTypes["map"] = tempOutNames;
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.valid(parameters, "outputdir"); if (outputDir == "not found"){ outputDir = ""; }
//check for required parameters
listfile = validParameter.validFile(parameters, "list");
if (listfile == "not open") { listfile = ""; abort = true; }
else if (listfile == "not found") { listfile = ""; }
else { current->setListFile(listfile); }
biomfile = validParameter.validFile(parameters, "biom");
if (biomfile == "not open") { biomfile = ""; abort = true; }
else if (biomfile == "not found") { biomfile = ""; }
else { current->setBiomFile(biomfile); }
ordergroupfile = validParameter.validFile(parameters, "ordergroup");
if (ordergroupfile == "not open") { abort = true; }
else if (ordergroupfile == "not found") { ordergroupfile = ""; }
groupfile = validParameter.validFile(parameters, "group");
if (groupfile == "not open") { groupfile = ""; abort = true; }
else if (groupfile == "not found") { groupfile = ""; }
else { current->setGroupFile(groupfile); }
countfile = validParameter.validFile(parameters, "count");
if (countfile == "not open") { countfile = ""; abort = true; }
else if (countfile == "not found") { countfile = ""; }
else {
current->setCountFile(countfile);
CountTable temp;
if (!temp.testGroups(countfile)) {
m->mothurOut("\n[WARNING]: Your count file does not have group info, all reads will be assigned to mothurGroup.\n");
temp.readTable(countfile, false, false); //dont read groups
map<string, int> seqs = temp.getNameMap();
CountTable newCountTable;
newCountTable.addGroup("mothurGroup");
for (map<string, int>::iterator it = seqs.begin(); it != seqs.end(); it++) {
vector<int> counts; counts.push_back(it->second);
newCountTable.push_back(it->first, counts);
}
string newCountfileName = util.getRootName(countfile) + "mothurGroup" + util.getExtension(countfile);
newCountTable.printTable(newCountfileName);
current->setCountFile(newCountfileName);
countfile = newCountfileName;
outputNames.push_back(newCountfileName);
}
}
if ((biomfile == "") && (listfile == "") && (countfile == "")) { //you must provide at least one of the following
//is there are current file available for either of these?
//give priority to list, then biom, then count
listfile = current->getListFile();
if (listfile != "") { m->mothurOut("Using " + listfile + " as input file for the list parameter.\n"); }
else {
biomfile = current->getBiomFile();
if (biomfile != "") { m->mothurOut("Using " + biomfile + " as input file for the biom parameter.\n"); }
else {
countfile = current->getCountFile();
if (countfile != "") { m->mothurOut("Using " + countfile + " as input file for the count parameter.\n"); }
else {
m->mothurOut("[ERROR]: No valid current files. You must provide a list or biom or count file before you can use the make.shared command.\n"); abort = true;
}
}
}
}
else if ((biomfile != "") && (listfile != "")) { m->mothurOut("When executing a make.shared command you must enter ONLY ONE of the following: list or biom.\n"); abort = true; }
if (listfile != "") {
if ((groupfile == "") && (countfile == "")) {
groupfile = current->getGroupFile();
if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter.\n"); }
else {
countfile = current->getCountFile();
if (countfile != "") { m->mothurOut("Using " + countfile + " as input file for the count parameter.\n"); }
else { m->mothurOut("[ERROR]: You need to provide a groupfile or countfile if you are going to use the list format.\n"); abort = true; }
}
}
}
string groups = validParameter.valid(parameters, "groups");
if (groups == "not found") { groups = ""; }
else {
pickedGroups=true;
util.splitAtDash(groups, Groups);
if (Groups.size() != 0) { if (Groups[0]== "all") { Groups.clear(); } }
}
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
string label = validParameter.valid(parameters, "label");
if (label == "not found") { label = ""; }
else {
if(label != "all") { util.splitAtDash(label, labels); allLines = 0; }
else { allLines = 1; }
}
if ((listfile == "") && (biomfile == "") && (countfile != "")) { //building a shared file from a count file, require label
if (labels.size() == 0) {
m->mothurOut("[ERROR]: You must provide a label when converting a count file to a shared file, please correct.\n"); abort = true;
}
}
}
}
catch(exception& e) {
m->errorOut(e, "SharedCommand", "SharedCommand");
exit(1);
}
}
int ClusterCommand::execute(){
try {
if (abort == true) { if (calledHelp) { return 0; } return 2; }
//phylip file given and cutoff not given - use cluster.classic because it uses less memory and is faster
if ((format == "phylip") && (cutoff > 10.0)) {
m->mothurOutEndLine(); m->mothurOut("You are using a phylip file and no cutoff. I will run cluster.classic to save memory and time."); m->mothurOutEndLine();
//run unique.seqs for deconvolute results
string inputString = "phylip=" + distfile;
if (namefile != "") { inputString += ", name=" + namefile; }
else if (countfile != "") { inputString += ", count=" + countfile; }
inputString += ", precision=" + toString(precision);
inputString += ", method=" + method;
if (hard) { inputString += ", hard=T"; }
else { inputString += ", hard=F"; }
if (sim) { inputString += ", sim=T"; }
else { inputString += ", sim=F"; }
m->mothurOutEndLine();
m->mothurOut("/------------------------------------------------------------/"); m->mothurOutEndLine();
m->mothurOut("Running command: cluster.classic(" + inputString + ")"); m->mothurOutEndLine();
Command* clusterClassicCommand = new ClusterDoturCommand(inputString);
clusterClassicCommand->execute();
delete clusterClassicCommand;
m->mothurOut("/------------------------------------------------------------/"); m->mothurOutEndLine();
return 0;
}
ReadMatrix* read;
if (format == "column") { read = new ReadColumnMatrix(columnfile, sim); } //sim indicates whether its a similarity matrix
else if (format == "phylip") { read = new ReadPhylipMatrix(phylipfile, sim); }
read->setCutoff(cutoff);
NameAssignment* nameMap = NULL;
CountTable* ct = NULL;
map<string, int> counts;
if(namefile != ""){
nameMap = new NameAssignment(namefile);
nameMap->readMap();
read->read(nameMap);
}else if (countfile != "") {
ct = new CountTable();
ct->readTable(countfile, false, false);
read->read(ct);
counts = ct->getNameMap();
}else { read->read(nameMap); }
list = read->getListVector();
matrix = read->getDMatrix();
if(countfile != "") {
rabund = new RAbundVector();
createRabund(ct, list, rabund); //creates an rabund that includes the counts for the unique list
delete ct;
}else { rabund = new RAbundVector(list->getRAbundVector()); }
delete read;
if (m->control_pressed) { //clean up
delete list; delete matrix; delete rabund; if(countfile == ""){rabundFile.close(); sabundFile.close(); m->mothurRemove((fileroot+ tag + ".rabund")); m->mothurRemove((fileroot+ tag + ".sabund")); }
listFile.close(); m->mothurRemove((fileroot+ tag + ".list")); outputTypes.clear(); return 0;
}
//create cluster
if (method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix, cutoff, method, adjust); }
else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix, cutoff, method, adjust); }
else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix, cutoff, method, adjust); }
else if(method == "weighted"){ cluster = new WeightedLinkage(rabund, list, matrix, cutoff, method, adjust); }
tag = cluster->getTag();
if (outputDir == "") { outputDir += m->hasPath(distfile); }
fileroot = outputDir + m->getRootName(m->getSimpleName(distfile));
map<string, string> variables;
variables["[filename]"] = fileroot;
variables["[clustertag]"] = tag;
string sabundFileName = getOutputFileName("sabund", variables);
string rabundFileName = getOutputFileName("rabund", variables);
if (countfile != "") { variables["[tag2]"] = "unique_list"; }
string listFileName = getOutputFileName("list", variables);
if (countfile == "") {
m->openOutputFile(sabundFileName, sabundFile);
m->openOutputFile(rabundFileName, rabundFile);
outputNames.push_back(sabundFileName); outputTypes["sabund"].push_back(sabundFileName);
outputNames.push_back(rabundFileName); outputTypes["rabund"].push_back(rabundFileName);
}
m->openOutputFile(listFileName, listFile);
outputNames.push_back(listFileName); outputTypes["list"].push_back(listFileName);
list->printHeaders(listFile);
time_t estart = time(NULL);
float previousDist = 0.00000;
float rndPreviousDist = 0.00000;
oldRAbund = *rabund;
oldList = *list;
print_start = true;
start = time(NULL);
loops = 0;
double saveCutoff = cutoff;
while (matrix->getSmallDist() < cutoff && matrix->getNNodes() > 0){
if (m->control_pressed) { //clean up
delete list; delete matrix; delete rabund; delete cluster;
if(countfile == "") {rabundFile.close(); sabundFile.close(); m->mothurRemove((fileroot+ tag + ".rabund")); m->mothurRemove((fileroot+ tag + ".sabund")); }
listFile.close(); m->mothurRemove((fileroot+ tag + ".list")); outputTypes.clear(); return 0;
}
if (print_start && m->isTrue(timing)) {
m->mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/"
+ toString(m->roundDist(matrix->getSmallDist(), precision))
+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
cout.flush();
print_start = false;
}
loops++;
cluster->update(cutoff);
float dist = matrix->getSmallDist();
float rndDist;
if (hard) {
rndDist = m->ceilDist(dist, precision);
}else{
rndDist = m->roundDist(dist, precision);
}
if(previousDist <= 0.0000 && dist != previousDist){
printData("unique", counts);
}
else if(rndDist != rndPreviousDist){
printData(toString(rndPreviousDist, length-1), counts);
}
previousDist = dist;
rndPreviousDist = rndDist;
oldRAbund = *rabund;
oldList = *list;
}
if (print_start && m->isTrue(timing)) {
m->mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist)
+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
cout.flush();
print_start = false;
}
if(previousDist <= 0.0000){
printData("unique", counts);
}
else if(rndPreviousDist<cutoff){
printData(toString(rndPreviousDist, length-1), counts);
}
delete matrix;
delete list;
delete rabund;
delete cluster;
if (countfile == "") {
sabundFile.close();
rabundFile.close();
}
listFile.close();
if (saveCutoff != cutoff) {
if (hard) { saveCutoff = m->ceilDist(saveCutoff, precision); }
else { saveCutoff = m->roundDist(saveCutoff, precision); }
m->mothurOut("changed cutoff to " + toString(cutoff)); m->mothurOutEndLine();
}
//set list file as new current listfile
string current = "";
itTypes = outputTypes.find("list");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); }
}
//set rabund file as new current rabundfile
itTypes = outputTypes.find("rabund");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRabundFile(current); }
}
//set sabund file as new current sabundfile
itTypes = outputTypes.find("sabund");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSabundFile(current); }
}
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
m->mothurOutEndLine();
//if (m->isTrue(timing)) {
m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine();
//}
return 0;
}
catch(exception& e) {
m->errorOut(e, "ClusterCommand", "execute");
exit(1);
}
}
int SeqSummaryCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
int start = time(NULL);
//set current fasta to fastafile
m->setFastaFile(fastafile);
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile));
string summaryFile = getOutputFileName("summary",variables);
long long numSeqs = 0;
long long size = 0;
long long numUniques = 0;
map<int, long long> startPosition;
map<int, long long> endPosition;
map<int, long long> seqLength;
map<int, long long> ambigBases;
map<int, long long> longHomoPolymer;
if (namefile != "") { nameMap = m->readNames(namefile); numUniques = nameMap.size(); }
else if (countfile != "") {
CountTable ct;
ct.readTable(countfile, false, false);
nameMap = ct.getNameMap();
size = ct.getNumSeqs();
numUniques = ct.getNumUniqueSeqs();
}
if (m->control_pressed) { return 0; }
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(fastafile, processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); }
#else
positions = m->setFilePosFasta(fastafile, numSeqs);
if (numSeqs < processors) { processors = numSeqs; }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor));
}
#endif
if(processors == 1){
numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile, lines[0]);
}else{
numSeqs = createProcessesCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile);
}
if (m->control_pressed) { return 0; }
//set size
if (countfile != "") {}//already set
else if (namefile == "") { size = numSeqs; }
else { for (map<int, long long>::iterator it = startPosition.begin(); it != startPosition.end(); it++) { size += it->second; } }
if ((namefile != "") || (countfile != "")) {
string type = "count";
if (namefile != "") { type = "name"; }
if (numSeqs != numUniques) { // do fasta and name/count files match
m->mothurOut("[ERROR]: Your " + type + " file contains " + toString(numUniques) + " unique sequences, but your fasta file contains " + toString(numSeqs) + ". File mismatch detected, quitting command.\n"); m->control_pressed = true;
}
}
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
long long ptile0_25 = 1+(long long)(size * 0.025); //number of sequences at 2.5%
long long ptile25 = 1+(long long)(size * 0.250); //number of sequences at 25%
long long ptile50 = 1+(long long)(size * 0.500);
long long ptile75 = 1+(long long)(size * 0.750);
long long ptile97_5 = 1+(long long)(size * 0.975);
long long ptile100 = (long long)(size);
vector<int> starts; starts.resize(7,0); vector<int> ends; ends.resize(7,0); vector<int> ambigs; ambigs.resize(7,0); vector<int> lengths; lengths.resize(7,0); vector<int> homops; homops.resize(7,0);
//find means
long long meanStartPosition, meanEndPosition, meanSeqLength, meanAmbigBases, meanLongHomoPolymer;
meanStartPosition = 0; meanEndPosition = 0; meanSeqLength = 0; meanAmbigBases = 0; meanLongHomoPolymer = 0;
//minimum
if ((startPosition.begin())->first == -1) { starts[0] = 0; }
else {starts[0] = (startPosition.begin())->first; }
long long totalSoFar = 0;
//set all values to min
starts[1] = starts[0]; starts[2] = starts[0]; starts[3] = starts[0]; starts[4] = starts[0]; starts[5] = starts[0];
int lastValue = 0;
for (map<int, long long>::iterator it = startPosition.begin(); it != startPosition.end(); it++) {
int value = it->first; if (value == -1) { value = 0; }
meanStartPosition += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ starts[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { starts[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { starts[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { starts[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { starts[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { starts[6] = value; } //save value
lastValue = totalSoFar;
}
starts[6] = (startPosition.rbegin())->first;
if ((endPosition.begin())->first == -1) { ends[0] = 0; }
else {ends[0] = (endPosition.begin())->first; }
totalSoFar = 0;
//set all values to min
ends[1] = ends[0]; ends[2] = ends[0]; ends[3] = ends[0]; ends[4] = ends[0]; ends[5] = ends[0];
lastValue = 0;
for (map<int, long long>::iterator it = endPosition.begin(); it != endPosition.end(); it++) {
int value = it->first; if (value == -1) { value = 0; }
meanEndPosition += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ ends[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { ends[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { ends[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { ends[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { ends[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { ends[6] = value; } //save value
lastValue = totalSoFar;
}
ends[6] = (endPosition.rbegin())->first;
if ((seqLength.begin())->first == -1) { lengths[0] = 0; }
else {lengths[0] = (seqLength.begin())->first; }
//set all values to min
lengths[1] = lengths[0]; lengths[2] = lengths[0]; lengths[3] = lengths[0]; lengths[4] = lengths[0]; lengths[5] = lengths[0];
totalSoFar = 0;
lastValue = 0;
for (map<int, long long>::iterator it = seqLength.begin(); it != seqLength.end(); it++) {
int value = it->first;
meanSeqLength += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ lengths[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { lengths[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { lengths[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { lengths[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { lengths[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { lengths[6] = value; } //save value
lastValue = totalSoFar;
}
lengths[6] = (seqLength.rbegin())->first;
if ((ambigBases.begin())->first == -1) { ambigs[0] = 0; }
else {ambigs[0] = (ambigBases.begin())->first; }
//set all values to min
ambigs[1] = ambigs[0]; ambigs[2] = ambigs[0]; ambigs[3] = ambigs[0]; ambigs[4] = ambigs[0]; ambigs[5] = ambigs[0];
totalSoFar = 0;
lastValue = 0;
for (map<int, long long>::iterator it = ambigBases.begin(); it != ambigBases.end(); it++) {
int value = it->first;
meanAmbigBases += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ ambigs[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { ambigs[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { ambigs[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { ambigs[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { ambigs[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { ambigs[6] = value; } //save value
lastValue = totalSoFar;
}
ambigs[6] = (ambigBases.rbegin())->first;
if ((longHomoPolymer.begin())->first == -1) { homops[0] = 0; }
else {homops[0] = (longHomoPolymer.begin())->first; }
//set all values to min
homops[1] = homops[0]; homops[2] = homops[0]; homops[3] = homops[0]; homops[4] = homops[0]; homops[5] = homops[0];
totalSoFar = 0;
lastValue = 0;
for (map<int, long long>::iterator it = longHomoPolymer.begin(); it != longHomoPolymer.end(); it++) {
int value = it->first;
meanLongHomoPolymer += (it->first*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ homops[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { homops[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { homops[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { homops[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { homops[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { homops[6] = value; } //save value
lastValue = totalSoFar;
}
homops[6] = (longHomoPolymer.rbegin())->first;
double meanstartPosition, meanendPosition, meanseqLength, meanambigBases, meanlongHomoPolymer;
meanstartPosition = meanStartPosition / (double) size; meanendPosition = meanEndPosition /(double) size; meanlongHomoPolymer = meanLongHomoPolymer / (double) size; meanseqLength = meanSeqLength / (double) size; meanambigBases = meanAmbigBases /(double) size;
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("\t\tStart\tEnd\tNBases\tAmbigs\tPolymer\tNumSeqs"); m->mothurOutEndLine();
m->mothurOut("Minimum:\t" + toString(starts[0]) + "\t" + toString(ends[0]) + "\t" + toString(lengths[0]) + "\t" + toString(ambigs[0]) + "\t" + toString(homops[0]) + "\t" + toString(1)); m->mothurOutEndLine();
m->mothurOut("2.5%-tile:\t" + toString(starts[1]) + "\t" + toString(ends[1]) + "\t" + toString(lengths[1]) + "\t" + toString(ambigs[1]) + "\t" + toString(homops[1]) + "\t" + toString(ptile0_25)); m->mothurOutEndLine();
m->mothurOut("25%-tile:\t" + toString(starts[2]) + "\t" + toString(ends[2]) + "\t" + toString(lengths[2]) + "\t" + toString(ambigs[2]) + "\t" + toString(homops[2]) + "\t" + toString(ptile25)); m->mothurOutEndLine();
m->mothurOut("Median: \t" + toString(starts[3]) + "\t" + toString(ends[3]) + "\t" + toString(lengths[3]) + "\t" + toString(ambigs[3]) + "\t" + toString(homops[3]) + "\t" + toString(ptile50)); m->mothurOutEndLine();
m->mothurOut("75%-tile:\t" + toString(starts[4]) + "\t" + toString(ends[4]) + "\t" + toString(lengths[4]) + "\t" + toString(ambigs[4]) + "\t" + toString(homops[4]) + "\t" + toString(ptile75)); m->mothurOutEndLine();
m->mothurOut("97.5%-tile:\t" + toString(starts[5]) + "\t" + toString(ends[5]) + "\t" + toString(lengths[5]) + "\t" + toString(ambigs[5]) + "\t" + toString(homops[5]) + "\t" + toString(ptile97_5)); m->mothurOutEndLine();
m->mothurOut("Maximum:\t" + toString(starts[6]) + "\t" + toString(ends[6]) + "\t" + toString(lengths[6]) + "\t" + toString(ambigs[6]) + "\t" + toString(homops[6]) + "\t" + toString(ptile100)); m->mothurOutEndLine();
m->mothurOut("Mean:\t" + toString(meanstartPosition) + "\t" + toString(meanendPosition) + "\t" + toString(meanseqLength) + "\t" + toString(meanambigBases) + "\t" + toString(meanlongHomoPolymer)); m->mothurOutEndLine();
if ((namefile == "") && (countfile == "")) { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); }
else { m->mothurOut("# of unique seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); m->mothurOut("total # of seqs:\t" + toString(size)); m->mothurOutEndLine(); }
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile);
m->mothurOutEndLine();
if ((namefile == "") && (countfile == "")) { m->mothurOut("It took " + toString(time(NULL) - start) + " secs to summarize " + toString(numSeqs) + " sequences.\n"); }
else{ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to summarize " + toString(size) + " sequences.\n"); }
//set fasta file as new current fastafile
string current = "";
itTypes = outputTypes.find("summary");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSummaryFile(current); }
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "SeqSummaryCommand", "execute");
exit(1);
}
}
int OptiBlastMatrix::readBlast(){
try {
Utils util;
map<string, long long> nameAssignment;
if (namefile != "") { util.readNames(namefile, nameAssignment); }
else if (countfile != "") {
CountTable ct; ct.readTable(countfile, false, true);
map<string, int> temp = ct.getNameMap();
for (map<string, int>::iterator it = temp.begin(); it!= temp.end(); it++) { nameAssignment[it->first] = it->second; }
}
else { readBlastNames(nameAssignment); }
int count = 0;
for (map<string, long long>::iterator it = nameAssignment.begin(); it!= nameAssignment.end(); it++) {
it->second = count; count++;
nameMap.push_back(it->first);
overlapNameMap.push_back(it->first);
}
m->mothurOut("Reading Blast File... "); cout.flush();
string firstName, secondName, eScore, currentRow; currentRow = "";
string repeatName = "";
float distance, thisoverlap, refScore;
float percentId;
float numBases, mismatch, gap, startQuery, endQuery, startRef, endRef, score, lengthThisSeq;
map<string, float> thisRowsBlastScores;
///////////////////// Read to eliminate singletons ///////////////////////
ifstream fileHandle;
util.openInputFile(distFile, fileHandle);
map<int, int> singletonIndexSwap;
map<int, int> blastSingletonIndexSwap;
vector<bool> singleton; singleton.resize(nameAssignment.size(), true);
vector<bool> overlapSingleton; overlapSingleton.resize(nameAssignment.size(), true);
vector< map<string,float> > dists; dists.resize(nameAssignment.size());
if (!fileHandle.eof()) {
//read in line from file
fileHandle >> firstName >> secondName >> percentId >> numBases >> mismatch >> gap >> startQuery >> endQuery >> startRef >> endRef >> eScore >> score;
util.gobble(fileHandle);
currentRow = firstName;
lengthThisSeq = numBases;
repeatName = firstName + secondName;
if (firstName == secondName) { refScore = score; }
else{
thisRowsBlastScores[secondName] = score;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
int indexA = (itA->second);
int indexB = (itB->second);
overlapSingleton[indexA] = false;
overlapSingleton[indexB] = false;
blastSingletonIndexSwap[indexA] = indexA;
blastSingletonIndexSwap[indexB] = indexB;
}
}
}else { m->mothurOut("Error in your blast file, cannot read."); m->mothurOutEndLine(); exit(1); }
while(fileHandle){ //let's assume it's a triangular matrix...
if (m->getControl_pressed()) { fileHandle.close(); return 0; }
//read in line from file
fileHandle >> firstName >> secondName >> percentId >> numBases >> mismatch >> gap >> startQuery >> endQuery >> startRef >> endRef >> eScore >> score;
util.gobble(fileHandle);
string temp = firstName + secondName; //to check if this file has repeat lines, ie. is this a blast instead of a blscreen file
//if this is a new pairing
if (temp != repeatName) {
repeatName = temp;
if (currentRow == firstName) {
if (firstName == secondName) { refScore = score; }
else{
//save score
thisRowsBlastScores[secondName] = score;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
int indexA = (itA->second);
int indexB = (itB->second);
overlapSingleton[indexA] = false;
overlapSingleton[indexB] = false;
blastSingletonIndexSwap[indexA] = indexA;
blastSingletonIndexSwap[indexB] = indexB;
}
} //end else
}else { //end row
//convert blast scores to distance and add cell to sparse matrix if we can
map<string, float>::iterator it;
map<string, float>::iterator itDist;
for(it=thisRowsBlastScores.begin(); it!=thisRowsBlastScores.end(); it++) {
distance = 1.0 - (it->second / refScore);
//do we already have the distance calculated for b->a
map<string,long long>::iterator itA = nameAssignment.find(currentRow);
map<string,long long>::iterator itB = nameAssignment.find(it->first);
itDist = dists[itB->second].find(itA->first);
//if we have it then compare
if (itDist != dists[itB->second].end()) {
//if you want the minimum blast score ratio, then pick max distance
if(minWanted) { distance = max(itDist->second, distance); }
else{ distance = min(itDist->second, distance); }
//is this distance below cutoff
if (distance <= cutoff) {
int indexA = (itA->second);
int indexB = (itB->second);
singleton[indexA] = false;
singleton[indexB] = false;
singletonIndexSwap[indexA] = indexA;
singletonIndexSwap[indexB] = indexB;
}
//not going to need this again
dists[itB->second].erase(itDist);
}else { //save this value until we get the other ratio
dists[itA->second][it->first] = distance;
}
}
//clear out last rows info
thisRowsBlastScores.clear();
currentRow = firstName;
lengthThisSeq = numBases;
//add this row to thisRowsBlastScores
if (firstName == secondName) { refScore = score; }
else{ //add this row to thisRowsBlastScores
thisRowsBlastScores[secondName] = score;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
int indexA = (itA->second);
int indexB = (itB->second);
overlapSingleton[indexA] = false;
overlapSingleton[indexB] = false;
blastSingletonIndexSwap[indexA] = indexA;
blastSingletonIndexSwap[indexB] = indexB;
}
}
}//end if current row
}//end if repeat
}
fileHandle.close();
//convert blast scores to distance and add cell to sparse matrix if we can
map<string, float>::iterator it;
map<string, float>::iterator itDist;
for(it=thisRowsBlastScores.begin(); it!=thisRowsBlastScores.end(); it++) {
distance = 1.0 - (it->second / refScore);
//do we already have the distance calculated for b->a
map<string,long long>::iterator itA = nameAssignment.find(currentRow);
map<string,long long>::iterator itB = nameAssignment.find(it->first);
itDist = dists[itB->second].find(itA->first);
//if we have it then compare
if (itDist != dists[itB->second].end()) {
//if you want the minimum blast score ratio, then pick max distance
if(minWanted) { distance = max(itDist->second, distance); }
else{ distance = min(itDist->second, distance); }
//is this distance below cutoff
if (distance <= cutoff) {
int indexA = (itA->second);
int indexB = (itB->second);
singleton[indexA] = false;
singleton[indexB] = false;
singletonIndexSwap[indexA] = indexA;
singletonIndexSwap[indexB] = indexB;
}
//not going to need this again
dists[itB->second].erase(itDist);
}else { //save this value until we get the other ratio
dists[itA->second][it->first] = distance;
}
}
//clear out info
thisRowsBlastScores.clear();
dists.clear();
//////////////////////////////////////////////////////////////////////////
int nonSingletonCount = 0;
for (int i = 0; i < singleton.size(); i++) {
if (!singleton[i]) { //if you are a singleton
singletonIndexSwap[i] = nonSingletonCount;
nonSingletonCount++;
}else { singletons.push_back(nameMap[i]); }
}
singleton.clear();
int overlapNonSingletonCount = 0;
for (int i = 0; i < overlapSingleton.size(); i++) {
if (!overlapSingleton[i]) { //if you are a singleton
blastSingletonIndexSwap[i] = overlapNonSingletonCount;
overlapNonSingletonCount++;
}
}
overlapSingleton.clear();
ifstream in;
util.openInputFile(distFile, in);
dists.resize(nameAssignment.size());
closeness.resize(nonSingletonCount);
blastOverlap.resize(overlapNonSingletonCount);
map<string, string> names;
if (namefile != "") {
util.readNames(namefile, names);
for (int i = 0; i < singletons.size(); i++) {
singletons[i] = names[singletons[i]];
}
}
m->mothurOut(" halfway ... "); cout.flush();
if (!in.eof()) {
//read in line from file
in >> firstName >> secondName >> percentId >> numBases >> mismatch >> gap >> startQuery >> endQuery >> startRef >> endRef >> eScore >> score;
util.gobble(fileHandle);
currentRow = firstName;
lengthThisSeq = numBases;
repeatName = firstName + secondName;
if (firstName == secondName) { refScore = score; }
else{
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
thisRowsBlastScores[secondName] = score;
if (namefile != "") {
firstName = names[firstName]; //redundant names
secondName = names[secondName]; //redundant names
}
nameMap[singletonIndexSwap[itA->second]] = firstName;
nameMap[singletonIndexSwap[itB->second]] = secondName;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
int indexA = (itA->second);
int indexB = (itB->second);
int newB = blastSingletonIndexSwap[indexB];
int newA = blastSingletonIndexSwap[indexA];
blastOverlap[newA].insert(newB);
blastOverlap[newB].insert(newA);
overlapNameMap[newA] = firstName;
overlapNameMap[newB] = secondName;
}
}
}else { m->mothurOut("Error in your blast file, cannot read."); m->mothurOutEndLine(); exit(1); }