#include "sequence.hpp"
#include "bayesian.h"
#include "phylotree.h"
+#include "phylosummary.h"
#include "knn.h"
//**********************************************************************************************************************
else {
//valid paramters for this command
- string AlignArray[] = {"template","fasta","name","search","ksize","method","processors","taxonomy","match","mismatch","gapopen","gapextend","numwanted","cutoff","probs","iters", "outputdir","inputdir"};
+ string AlignArray[] = {"template","fasta","name","group","search","ksize","method","processors","taxonomy","match","mismatch","gapopen","gapextend","numwanted","cutoff","probs","iters", "outputdir","inputdir"};
vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
OptionParser parser(option);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
}
+
+ it = parameters.find("group");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = 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; }
+ }
}
//check for required parameters
}
else if (templateFileName == "not open") { abort = true; }
+ groupfile = validParameter.validFile(parameters, "group", true);
+ if (groupfile == "not open") { abort = true; }
+ else if (groupfile == "not found") { groupfile = ""; }
+
fastaFileName = validParameter.validFile(parameters, "fasta", false);
if (fastaFileName == "not found") { m->mothurOut("fasta is a required parameter for the classify.seqs command."); m->mothurOutEndLine(); abort = true; }
else {
m->mothurOut("The template, fasta and taxonomy parameters are required. You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n");
m->mothurOut("The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer, blast and distance. The default is kmer.\n");
m->mothurOut("The name parameter allows you add a names file with your fasta file, if you enter multiple fasta files, you must enter matching names files for them.\n");
+ m->mothurOut("The group parameter allows you add a group file so you can have the summary totals broken up by group.\n");
m->mothurOut("The method parameter allows you to specify classification method to use. Your options are: bayesian and knn. The default is bayesian.\n");
m->mothurOut("The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 8.\n");
m->mothurOut("The processors parameter allows you to specify the number of processors to use. The default is 1.\n");
int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
int inMode=MPI_MODE_RDONLY;
-
- char outNewTax[newTaxonomyFile.length()];
+
+ //char* outNewTax = new char[newTaxonomyFile.length()];
+ //memcpy(outNewTax, newTaxonomyFile.c_str(), newTaxonomyFile.length());
+
+ char outNewTax[1024];
strcpy(outNewTax, newTaxonomyFile.c_str());
+
+ //char* outTempTax = new char[tempTaxonomyFile.length()];
+ //memcpy(outTempTax, tempTaxonomyFile.c_str(), tempTaxonomyFile.length());
- char outTempTax[tempTaxonomyFile.length()];
+ char outTempTax[1024];
strcpy(outTempTax, tempTaxonomyFile.c_str());
+
+ //char* inFileName = new char[fastaFileNames[s].length()];
+ //memcpy(inFileName, fastaFileNames[s].c_str(), fastaFileNames[s].length());
- char inFileName[fastaFileNames[s].length()];
+ char inFileName[1024];
strcpy(inFileName, fastaFileNames[s].c_str());
MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
MPI_File_open(MPI_COMM_WORLD, outNewTax, outMode, MPI_INFO_NULL, &outMPINewTax);
MPI_File_open(MPI_COMM_WORLD, outTempTax, outMode, MPI_INFO_NULL, &outMPITempTax);
+ //delete outNewTax;
+ //delete outTempTax;
+ //delete inFileName;
+
if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
if(namefile != "") { MPIReadNamesFile(namefileNames[s]); }
if (pid == 0) { //this part does not need to be paralellized
#endif
- //make taxonomy tree from new taxonomy file
- PhyloTree taxaBrowser;
+ m->mothurOutEndLine();
+ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
+ start = time(NULL);
+
+ PhyloSummary taxaSum(taxonomyFileName, groupfile);
if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
- ifstream in;
- openInputFile(tempTaxonomyFile, in);
-
- //read in users taxonomy file and add sequences to tree
- string name, taxon;
- while(!in.eof()){
- in >> name >> taxon; gobble(in);
-
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(tempTaxonomyFile.c_str()); delete classify; return 0; }
+ if (namefile == "") { taxaSum.summarize(tempTaxonomyFile); }
+ else {
+ ifstream in;
+ openInputFile(tempTaxonomyFile, in);
- if (namefile != "") {
+ //read in users taxonomy file and add sequences to tree
+ string name, taxon;
+ while(!in.eof()){
+ in >> name >> taxon; gobble(in);
+
itNames = nameMap.find(name);
if (itNames == nameMap.end()) {
m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1);
}else{
for (int i = 0; i < itNames->second; i++) {
- taxaBrowser.addSeqToTree(name+toString(i), taxon); //add it as many times as there are identical seqs
+ taxaSum.addSeqToTree(name+toString(i), taxon); //add it as many times as there are identical seqs
}
}
- }else { taxaBrowser.addSeqToTree(name, taxon); } //add it once
+ }
+ in.close();
}
- in.close();
-
- taxaBrowser.assignHeirarchyIDs(0);
-
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(tempTaxonomyFile.c_str()); delete classify; return 0; }
-
- taxaBrowser.binUnclassified();
-
remove(tempTaxonomyFile.c_str());
if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
-
//print summary file
ofstream outTaxTree;
openOutputFile(taxSummary, outTaxTree);
- taxaBrowser.print(outTaxTree);
+ taxaSum.print(outTaxTree);
outTaxTree.close();
//output taxonomy with the unclassified bins added
openOutputFile(unclass, outTax);
//get maxLevel from phylotree so you know how many 'unclassified's to add
- int maxLevel = taxaBrowser.getMaxLevel();
+ int maxLevel = taxaSum.getMaxLevel();
//read taxfile - this reading and rewriting is done to preserve the confidence scores.
+ string name, taxon;
while (!inTax.eof()) {
if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(unclass.c_str()); delete classify; return 0; }
remove(newTaxonomyFile.c_str());
rename(unclass.c_str(), newTaxonomyFile.c_str());
+ m->mothurOutEndLine();
+ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
+
#ifdef USE_MPI
}
#endif
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
m->mothurOutEndLine();
-
- m->mothurOutEndLine();
- m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
}
delete classify;
//read next sequence
int length = MPIPos[start+i+1] - MPIPos[start+i];
- char buf4[length];
+ char* buf4 = new char[length];
MPI_File_read_at(inMPI, MPIPos[start+i], buf4, length, MPI_CHAR, &status);
string tempBuf = buf4;
if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
istringstream iss (tempBuf,istringstream::in);
+ delete buf4;
Sequence* candidateSeq = new Sequence(iss);
}
int length = outputString.length();
- char buf2[length];
- strcpy(buf2, outputString.c_str());
+ char* buf2 = new char[length];
+ memcpy(buf2, outputString.c_str(), length);
MPI_File_write_shared(newFile, buf2, length, MPI_CHAR, &statusNew);
-
+ delete buf2;
+
outputString = candidateSeq->getName() + "\t" + classify->getSimpleTax() + "\n";
length = outputString.length();
- char buf[length];
- strcpy(buf, outputString.c_str());
+ char* buf = new char[length];
+ memcpy(buf, outputString.c_str(), length);
MPI_File_write_shared(tempFile, buf, length, MPI_CHAR, &statusTemp);
+ delete buf;
}
}
delete candidateSeq;
MPI_File inMPI;
MPI_Offset size;
MPI_Status status;
+
+ //char* inFileName = new char[nameFilename.length()];
+ //memcpy(inFileName, nameFilename.c_str(), nameFilename.length());
- char inFileName[nameFilename.length()];
+ char inFileName[1024];
strcpy(inFileName, nameFilename.c_str());
MPI_File_open(MPI_COMM_WORLD, inFileName, MPI_MODE_RDONLY, MPI_INFO_NULL, &inMPI);
MPI_File_get_size(inMPI, &size);
+ //delete inFileName;
- char buffer[size];
+ char* buffer = new char[size];
MPI_File_read(inMPI, buffer, size, MPI_CHAR, &status);
string tempBuf = buffer;
if (tempBuf.length() > size) { tempBuf = tempBuf.substr(0, size); }
istringstream iss (tempBuf,istringstream::in);
+ delete buffer;
string firstCol, secondCol;
while(!iss.eof()) {