vector<string> ShhherCommand::getValidParameters(){
try {
string Array[] = {
- "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors", "maxiter", "mindelta"
+ "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors", "maxiter", "mindelta", "order"
};
vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
//valid paramters for this command
string AlignArray[] = {
- "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors", "maxiter", "mindelta"
+ "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors", "maxiter", "mindelta", "order"
};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
temp = validParameter.validFile(parameters, "sigma", false);if (temp == "not found") { temp = "60"; }
convert(temp, sigma);
+ flowOrder = validParameter.validFile(parameters, "order", false);
+ if (flowOrder == "not found"){ flowOrder = "TACG"; }
+ else if(flowOrder.length() != 4){
+ m->mothurOut("The value of the order option must be four bases long\n");
+ }
+
globaldata = GlobalData::getInstance();
}
int tag = 1976;
MPI_Status status;
- double begClock = clock();
- unsigned long int begTime = time(NULL);
-
if(pid == 0){
for(int i=1;i<ncpus;i++){
}
for(int i=0;i<numFiles;i++){
- flowFileName = flowFileVector[i];
+ double begClock = clock();
+ unsigned long int begTime = time(NULL);
-
+ flowFileName = flowFileVector[i];
m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n");
m->mothurOut("Reading flowgrams...\n");
MPI_Send(&nSeqsPerOTU[0], numOTUs, MPI_INT, i, tag, MPI_COMM_WORLD);
MPI_Send(&cumNumSeqs[0], numOTUs, MPI_INT, i, tag, MPI_COMM_WORLD);
}
-
+
calcCentroidsDriver(0, numOTUsOnCPU);
for(int i=1;i<ncpus;i++){
}
else{
int abort = 1;
- bool live = 1;
MPI_Recv(&abort, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
if(abort){ return 0; }
for(int i=0;i<numFiles;i++){
//Now into the pyrodist part
+ bool live = 1;
+
char fileName[1024];
MPI_Recv(&fileName, 1024, MPI_CHAR, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
int total;
while(live){
-
+
MPI_Recv(&total, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
singleTau.assign(total, 0.0000);
seqNumber.assign(total, 0);
seqIndex.assign(total, 0);
-
+
MPI_Recv(&change[0], numOTUs, MPI_SHORT, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(¢roids[0], numOTUs, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&singleTau[0], total, MPI_DOUBLE, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&seqIndex[0], total, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&nSeqsPerOTU[0], total, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&cumNumSeqs[0], numOTUs, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
-
+
calcCentroidsDriver(startOTU, endOTU);
MPI_Send(¢roids[0], numOTUs, MPI_INT, 0, tag, MPI_COMM_WORLD);
MPI_Send(&change[0], numOTUs, MPI_SHORT, 0, tag, MPI_COMM_WORLD);
-
MPI_Recv(¢roids[0], numOTUs, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&weight[0], numOTUs, MPI_DOUBLE, 0, tag, MPI_COMM_WORLD, &status);
MPI_Recv(&change[0], numOTUs, MPI_SHORT, 0, tag, MPI_COMM_WORLD, &status);
}
}
}
-
MPI_Barrier(MPI_COMM_WORLD);
+
return 0;
}
double begClock = clock();
unsigned long int begTime = time(NULL);
-
m->mothurOut("\nDenoising flowgrams...\n");
m->mothurOut("iter\tmaxDelta\tnLL\t\tcycletime\n");
void ShhherCommand::writeSequences(vector<int> otuCounts){
try {
- string bases = "TACG";
string fastaFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".pn.fasta";
ofstream fastaFile;
if(otuCounts[i] > 0){
fastaFile << '>' << seqNameVector[aaI[i][0]] << endl;
- for(int j=8;j<numFlowCells;j++){
+ string newSeq = "";
+
+ for(int j=0;j<numFlowCells;j++){
- char base = bases[j % 4];
+ char base = flowOrder[j % 4];
for(int k=0;k<uniqueFlowgrams[index * numFlowCells + j];k++){
- fastaFile << base;
+ newSeq += base;
}
}
- fastaFile << endl;
+
+ fastaFile << newSeq.substr(4) << endl;
}
}
fastaFile.close();
ofstream otuCountsFile;
m->openOutputFile(otuCountsFileName, otuCountsFile);
- string bases = "TACG";
+ string bases = flowOrder;
for(int i=0;i<numOTUs;i++){
//output the translated version of the centroid sequence for the otu
int sequence = aaI[i][j];
otuCountsFile << seqNameVector[sequence] << '\t';
- for(int k=8;k<lengths[sequence];k++){
+ string newSeq = "";
+
+ for(int k=0;k<lengths[sequence];k++){
char base = bases[k % 4];
int freq = int(0.01 * (double)flowDataIntI[sequence * numFlowCells + k] + 0.5);
-
+
for(int s=0;s<freq;s++){
- otuCountsFile << base;
+ newSeq += base;
+ //otuCountsFile << base;
}
}
- otuCountsFile << endl;
+ otuCountsFile << newSeq.substr(4) << endl;
}
otuCountsFile << endl;
}