#include "listvector.hpp"
#include "cluster.hpp"
#include "sparsematrix.hpp"
+#include <cfloat>
//**********************************************************************************************************************
ShhherCommand::ShhherCommand(){
try {
- abort = true;
+ abort = true; calledHelp = true;
//initialize outputTypes
vector<string> tempOutNames;
#endif
- 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 {
m->mothurOutEndLine();
abort = true;
}
- else if (flowFileName == "not open" || flowFilesFileName == "not open") { abort = true; }
+ else if (flowFileName == "not open" || flowFilesFileName == "not open") { abort = true; }
+
+ if(flowFileName != "not found"){ compositeFASTAFileName = ""; }
+ else{
+ compositeFASTAFileName = flowFilesFileName.substr(0, flowFilesFileName.length()-10) + "pn.fasta";
+ ofstream temp;
+ m->openOutputFile(compositeFASTAFileName, temp);
+ temp.close();
+ }
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
// ...at some point should added some additional type checking...
string temp;
temp = validParameter.validFile(parameters, "lookup", true);
- if (temp == "not found") { lookupFileName = "LookUp_E123.pat"; }
+ if (temp == "not found") { lookupFileName = "LookUp_Titanium.pat"; }
else if(temp == "not open") { abort = true; }
else { lookupFileName = temp; }
#ifdef USE_MPI
int ShhherCommand::execute(){
try {
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
+
int tag = 1976;
MPI_Status status;
double begClock = clock();
unsigned long int begTime = time(NULL);
-
- cout.setf(ios::fixed, ios::floatfield);
- cout.setf(ios::showpoint);
- cout << setprecision(2);
-
if(pid == 0){
processors = ncpus;
- cout << "\nGetting preliminary data..." << endl;
+ m->mothurOut("\nGetting preliminary data...\n");
getSingleLookUp();
getJointLookUp();
-
+
vector<string> flowFileVector;
if(flowFilesFileName != "not found"){
string fName;
for(int i=0;i<numFiles;i++){
flowFileName = flowFileVector[i];
- cout << "\n>>>>>\tProcessing " << flowFileName << " (file " << i+1 << " of " << numFiles << ")\t<<<<<" << endl;
- cout << "Reading flowgrams..." << endl;
+ m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n");
+ m->mothurOut("Reading flowgrams...\n");
+
getFlowData();
- cout << "Identifying unique flowgrams..." << endl;
+ m->mothurOut("Identifying unique flowgrams...\n");
getUniques();
- cout << "Calculating distances between flowgrams..." << endl;
+ m->mothurOut("Calculating distances between flowgrams...\n");
char fileName[1024];
strcpy(fileName, flowFileName.c_str());
string namesFileName = createNamesFile();
- cout << "\nClustering flowgrams..." << endl;
+ m->mothurOut("\nClustering flowgrams...\n");
string listFileName = cluster(distFileName, namesFileName);
- // string listFileName = "PriestPot_C7.pn.list";
- // string listFileName = "test.mock_rep3.v69.pn.list";
-
+
getOTUData(listFileName);
initPyroCluster();
int numOTUsOnCPU = numOTUs / ncpus;
int numSeqsOnCPU = numSeqs / ncpus;
-
- cout << "\nDenoising flowgrams..." << endl;
- cout << "iter\tmaxDelta\tnLL\t\tcycletime" << endl;
+ m->mothurOut("\nDenoising flowgrams...\n");
+ m->mothurOut("iter\tmaxDelta\tnLL\t\tcycletime\n");
while((maxIters == 0 && maxDelta > minDelta) || iter < MIN_ITER || (maxDelta > minDelta && iter < maxIters)){
iter++;
- cout << iter << '\t' << maxDelta << '\t' << setprecision(2) << nLL << '\t' << time(NULL) - cycTime << '\t' << setprecision(6) << (clock() - cycClock)/(double)CLOCKS_PER_SEC << endl;
+ m->mothurOut(toString(iter) + '\t' + toString(maxDelta) + '\t' + toString(nLL) + '\t' + toString(time(NULL) - cycTime) + '\t' + toString((clock() - cycClock)/(double)CLOCKS_PER_SEC) + '\n');
if((maxIters == 0 && maxDelta > minDelta) || iter < MIN_ITER || (maxDelta > minDelta && iter < maxIters)){
int live = 1;
}
- cout << "\nFinalizing..." << endl;
+ m->mothurOut("\nFinalizing...\n");
fill();
setOTUs();
vector<int> otuCounts(numOTUs, 0);
remove(distFileName.c_str());
remove(namesFileName.c_str());
remove(listFileName.c_str());
-
- cout << "Total time to process " << flowFileName << ":\t" << time(NULL) - begTime << '\t' << setprecision(6) << (clock() - begClock)/(double)CLOCKS_PER_SEC << endl;
+
+ m->mothurOut("Total time to process " + toString(flowFileName) + ":\t" + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/(double)CLOCKS_PER_SEC) + '\n');
}
-
}
else{
int abort = 1;
}
}
if(i % 100 == 0){
- cout << i << "\t" << (time(NULL) - begTime) << "\t" << (clock()-begClock)/CLOCKS_PER_SEC << endl;
+ m->mothurOut(toString(i) + '\t' + toString(time(NULL) - begTime) + '\t' + toString((clock()-begClock)/CLOCKS_PER_SEC) + '\n');
}
}
- cout << stopSeq << "\t" << (time(NULL) - begTime) << "\t" << (clock()-begClock)/CLOCKS_PER_SEC << endl;
+
+ m->mothurOut(toString(stopSeq) + '\t' + toString(time(NULL) - begTime) + '\t' + toString((clock()-begClock)/CLOCKS_PER_SEC) + '\n');
string fDistFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".pn.dist";
if(pid != 0){ fDistFileName += ".temp." + toString(pid); }
try {
if (abort == true) { return 0; }
- cout.setf(ios::fixed, ios::floatfield);
- cout.setf(ios::showpoint);
-
getSingleLookUp();
getJointLookUp();
-
-
+
vector<string> flowFileVector;
if(flowFilesFileName != "not found"){
string fName;
for(int i=0;i<numFiles;i++){
flowFileName = flowFileVector[i];
- cout << "\n>>>>>\tProcessing " << flowFileName << " (file " << i+1 << " of " << numFiles << ")\t<<<<<" << endl;
- cout << "Reading flowgrams..." << endl;
+ m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n");
+ m->mothurOut("Reading flowgrams...\n");
getFlowData();
- cout << "Identifying unique flowgrams..." << endl;
+
+ m->mothurOut("Identifying unique flowgrams...\n");
getUniques();
- cout << "Calculating distances between flowgrams..." << endl;
+ m->mothurOut("Calculating distances between flowgrams...\n");
string distFileName = createDistFile(processors);
string namesFileName = createNamesFile();
-
- cout << "\nClustering flowgrams..." << endl;
+
+ m->mothurOut("\nClustering flowgrams...\n");
string listFileName = cluster(distFileName, namesFileName);
getOTUData(listFileName);
double begClock = clock();
unsigned long int begTime = time(NULL);
- cout << "\nDenoising flowgrams..." << endl;
- cout << "iter\tmaxDelta\tnLL\t\tcycletime" << endl;
+ m->mothurOut("\nDenoising flowgrams...\n");
+ m->mothurOut("iter\tmaxDelta\tnLL\t\tcycletime\n");
while((maxIters == 0 && maxDelta > minDelta) || iter < MIN_ITER || (maxDelta > minDelta && iter < maxIters)){
iter++;
- cout << iter << '\t' << maxDelta << '\t' << setprecision(2) << nLL << '\t' << time(NULL) - cycTime << '\t' << setprecision(6) << (clock() - cycClock)/(double)CLOCKS_PER_SEC << endl;
+ m->mothurOut(toString(iter) + '\t' + toString(maxDelta) + '\t' + toString(nLL) + '\t' + toString(time(NULL) - cycTime) + '\t' + toString((clock() - cycClock)/(double)CLOCKS_PER_SEC) + '\n');
+
}
- cout << "\nFinalizing..." << endl;
+ m->mothurOut("\nFinalizing...\n");
fill();
setOTUs();
remove(namesFileName.c_str());
remove(listFileName.c_str());
- cout << "Total time to process " << flowFileName << ":\t" << time(NULL) - begTime << '\t' << setprecision(6) << (clock() - begClock)/(double)CLOCKS_PER_SEC << endl;
+ m->mothurOut("Total time to process " + flowFileName + ":\t" + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/(double)CLOCKS_PER_SEC) + '\n');
}
return 0;
}
for(int i=0;i<NUMBINS;i++){
for(int j=0;j<NUMBINS;j++){
- float minSum = 10000000000;
+ double minSum = 100000000;
for(int k=0;k<HOMOPS;k++){
- float sum = singleLookUp[k * NUMBINS + i] + singleLookUp[k * NUMBINS + j];
+ double sum = singleLookUp[k * NUMBINS + i] + singleLookUp[k * NUMBINS + j];
if(sum < minSum) { minSum = sum; }
}
/**************************************************************************************************/
-float ShhherCommand::getProbIntensity(int intIntensity){
+double ShhherCommand::getProbIntensity(int intIntensity){
try{
- float minNegLogProb = 10000000000;
+
+ double minNegLogProb = 100000000;
+
for(int i=0;i<HOMOPS;i++){//loop signal strength
float negLogProb = singleLookUp[i * NUMBINS + intIntensity];
int index = 0;
vector<short> current(numFlowCells);
- for(int j=0;j<numFlowCells;j++){ current[j] = short(((flowDataIntI[i * numFlowCells + j] + 50.0)/100.0)); }
+ for(int j=0;j<numFlowCells;j++){
+ current[j] = short(((flowDataIntI[i * numFlowCells + j] + 50.0)/100.0));
+ }
for(int j=0;j<numUniques;j++){
int offset = j * numFlowCells;
bool toEnd = 1;
- for(int k=0;k<numFlowCells;k++){
+ int shorterLength;
+ if(lengths[i] < uniqueLengths[j]) { shorterLength = lengths[i]; }
+ else { shorterLength = uniqueLengths[j]; }
+
+ for(int k=0;k<shorterLength;k++){
if(current[k] != uniqueFlowgrams[offset + k]){
toEnd = 0;
break;
mapSeqToUnique[i] = j;
uniqueCount[j]++;
index = j;
+ if(lengths[i] > uniqueLengths[j]) { uniqueLengths[j] = lengths[i]; }
break;
}
index++;
float flowDistance = calcPairwiseDist(mapUniqueToSeq[i], mapUniqueToSeq[j]);
if(flowDistance < 1e-6){
- outStream << seqNameVector[mapUniqueToSeq[i]] << '\t' << seqNameVector[mapUniqueToSeq[j]] << '\t' << 0.000000 << endl;
+ outStream << mapUniqueToSeq[i] << '\t' << mapUniqueToSeq[j] << '\t' << 0.000000 << endl;
}
else if(flowDistance <= cutoff){
- outStream << seqNameVector[mapUniqueToSeq[i]] << '\t' << seqNameVector[mapUniqueToSeq[j]] << '\t' << flowDistance << endl;
+ outStream << mapUniqueToSeq[i] << '\t' << mapUniqueToSeq[j] << '\t' << flowDistance << endl;
}
}
if(i % 100 == 0){
m->mothurOutEndLine();
- cout << "Total time: " << (time(NULL) - begTime) << "\t" << (clock() - begClock)/CLOCKS_PER_SEC << endl;;
+ m->mothurOut("Total time: " + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/CLOCKS_PER_SEC) + '\n');
+
return fDistFileName;
}
string ShhherCommand::cluster(string distFileName, string namesFileName){
try {
- SparseMatrix* matrix;
- ListVector* list;
- RAbundVector* rabund;
globaldata->setNameFile(namesFileName);
globaldata->setColumnFile(distFileName);
clusterNameMap->readMap();
read->read(clusterNameMap);
- list = read->getListVector();
- matrix = read->getMatrix();
+ ListVector* list = read->getListVector();
+ SparseMatrix* matrix = read->getMatrix();
delete read;
delete clusterNameMap;
- rabund = new RAbundVector(list->getRAbundVector());
+ RAbundVector* rabund = new RAbundVector(list->getRAbundVector());
Cluster* cluster = new CompleteLinkage(rabund, list, matrix, cutoff, "furthest");
string tag = cluster->getTag();
double clusterCutoff = cutoff;
while (matrix->getSmallDist() <= clusterCutoff && matrix->getNNodes() > 0){
cluster->update(clusterCutoff);
- float dist = matrix->getSmallDist();
}
list->setLabel(toString(cutoff));
for(int j=0;j<nSeqsPerOTU[i];j++){
int index = cumNumSeqs[i] + j;
int nI = seqIndex[index];
- int nIU = mapSeqToUnique[nI];
double tauValue = singleTau[seqNumber[index]];
for(int j=0;j<nSeqsPerOTU[i];j++){
int index = cumNumSeqs[i] + j;
- int nI = seqIndex[index];
double tauValue = singleTau[seqNumber[index]];
weight[i] += tauValue;
}
try{
- vector<double> P(numSeqs, 0);
+ vector<long double> P(numSeqs, 0);
int effNumOTUs = 0;
for(int i=0;i<numOTUs;i++){
}
}
+ string hold;
for(int i=0;i<numOTUs;i++){
for(int j=0;j<nSeqsPerOTU[i];j++){
int index = cumNumSeqs[i] + j;
P[nI] += weight[i] * exp(-singleDist * sigma);
}
}
-
double nLL = 0.00;
for(int i=0;i<numSeqs;i++){
+ if(P[i] == 0){ P[i] = DBL_EPSILON; }
+
nLL += -log(P[i]);
}
nLL = nLL -(double)numSeqs * log(sigma);
-
+
return nLL;
}
catch(exception& e) {
vector<vector<int> > qualities(numOTUs);
vector<double> pr(HOMOPS, 0);
- int index = 0;
for(int i=0;i<numOTUs;i++){
int index = 0;
void ShhherCommand::writeSequences(vector<int> otuCounts){
try {
-
string bases = "TACG";
string fastaFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".pn.fasta";
}
}
fastaFile.close();
+
+ if(compositeFASTAFileName != ""){
+ m->appendFiles(fastaFileName, compositeFASTAFileName);
+ }
}
catch(exception& e) {
m->errorOut(e, "ShhherCommand", "writeSequences");