]> git.donarmstrong.com Git - mothur.git/blobdiff - bayesian.cpp
moved mothur's source into a folder to make grabbing just the source easier on github
[mothur.git] / bayesian.cpp
diff --git a/bayesian.cpp b/bayesian.cpp
deleted file mode 100644 (file)
index 1dc3833..0000000
+++ /dev/null
@@ -1,761 +0,0 @@
-/*
- *  bayesian.cpp
- *  Mothur
- *
- *  Created by westcott on 11/3/09.
- *  Copyright 2009 Schloss Lab. All rights reserved.
- *
- */
-
-#include "bayesian.h"
-#include "kmer.hpp"
-#include "phylosummary.h"
-#include "referencedb.h"
-/**************************************************************************************************/
-Bayesian::Bayesian(string tfile, string tempFile, string method, int ksize, int cutoff, int i, int tid, bool f) : 
-Classify(), kmerSize(ksize), confidenceThreshold(cutoff), iters(i) {
-       try {
-               ReferenceDB* rdb = ReferenceDB::getInstance();
-               
-               threadID = tid;
-               flip = f;
-               string baseName = tempFile;
-                       
-               if (baseName == "saved") { baseName = rdb->getSavedReference(); }
-               
-               string baseTName = tfile;
-               if (baseTName == "saved") { baseTName = rdb->getSavedTaxonomy(); }
-               
-               /************calculate the probablity that each word will be in a specific taxonomy*************/
-               string tfileroot = m->getFullPathName(baseTName.substr(0,baseTName.find_last_of(".")+1));
-               string tempfileroot = m->getRootName(m->getSimpleName(baseName));
-               string phyloTreeName = tfileroot + "tree.train";
-               string phyloTreeSumName = tfileroot + "tree.sum";
-               string probFileName = tfileroot + tempfileroot + char('0'+ kmerSize) + "mer.prob";
-               string probFileName2 = tfileroot + tempfileroot + char('0'+ kmerSize) + "mer.numNonZero";
-               
-               ofstream out;
-               ofstream out2;
-               
-               ifstream phyloTreeTest(phyloTreeName.c_str());
-               ifstream probFileTest2(probFileName2.c_str());
-               ifstream probFileTest(probFileName.c_str());
-               ifstream probFileTest3(phyloTreeSumName.c_str());
-               
-               int start = time(NULL);
-               
-               //if they are there make sure they were created after this release date
-               bool FilesGood = false;
-               if(probFileTest && probFileTest2 && phyloTreeTest && probFileTest3){
-                       FilesGood = checkReleaseDate(probFileTest, probFileTest2, phyloTreeTest, probFileTest3);
-               }
-               
-               //if you want to save, but you dont need to calculate then just read
-               if (rdb->save && probFileTest && probFileTest2 && phyloTreeTest && probFileTest3 && FilesGood && (tempFile != "saved")) {  
-                       ifstream saveIn;
-                       m->openInputFile(tempFile, saveIn);
-                       
-                       while (!saveIn.eof()) {
-                               Sequence temp(saveIn);
-                               m->gobble(saveIn);
-                               
-                               rdb->referenceSeqs.push_back(temp); 
-                       }
-                       saveIn.close();                 
-               }
-               
-               if(probFileTest && probFileTest2 && phyloTreeTest && probFileTest3 && FilesGood){       
-                       if (tempFile == "saved") { m->mothurOutEndLine();  m->mothurOut("Using sequences from " + rdb->getSavedReference() + " that are saved in memory.");     m->mothurOutEndLine(); }
-                       
-                       m->mothurOut("Reading template taxonomy...     "); cout.flush();
-                       
-                       phyloTree = new PhyloTree(phyloTreeTest, phyloTreeName);
-                       
-                       m->mothurOut("DONE."); m->mothurOutEndLine();
-                       
-                       genusNodes = phyloTree->getGenusNodes(); 
-                       genusTotals = phyloTree->getGenusTotals();
-                       
-                       if (tfile == "saved") { 
-                               m->mothurOutEndLine();  m->mothurOut("Using probabilties from " + rdb->getSavedTaxonomy() + " that are saved in memory...    ");        cout.flush();; 
-                               wordGenusProb = rdb->wordGenusProb;
-                               WordPairDiffArr = rdb->WordPairDiffArr;
-                       }else {
-                               m->mothurOut("Reading template probabilities...     "); cout.flush();
-                               readProbFile(probFileTest, probFileTest2, probFileName, probFileName2);
-                       }       
-                       
-                       //save probabilities
-                       if (rdb->save) { rdb->wordGenusProb = wordGenusProb; rdb->WordPairDiffArr = WordPairDiffArr; }
-               }else{
-               
-                       //create search database and names vector
-                       generateDatabaseAndNames(tfile, tempFile, method, ksize, 0.0, 0.0, 0.0, 0.0);
-                       
-                       //prevents errors caused by creating shortcut files if you had an error in the sanity check.
-                       if (m->control_pressed) {  m->mothurRemove(phyloTreeName);  m->mothurRemove(probFileName); m->mothurRemove(probFileName2); }
-                       else{ 
-                               genusNodes = phyloTree->getGenusNodes(); 
-                               genusTotals = phyloTree->getGenusTotals();
-                               
-                               m->mothurOut("Calculating template taxonomy tree...     "); cout.flush();
-                               
-                               phyloTree->printTreeNodes(phyloTreeName);
-                                                       
-                               m->mothurOut("DONE."); m->mothurOutEndLine();
-                               
-                               m->mothurOut("Calculating template probabilities...     "); cout.flush();
-                               
-                               numKmers = database->getMaxKmer() + 1;
-                       
-                               //initialze probabilities
-                               wordGenusProb.resize(numKmers);
-                               WordPairDiffArr.resize(numKmers);
-                       
-                               for (int j = 0; j < wordGenusProb.size(); j++) {        wordGenusProb[j].resize(genusNodes.size());             }
-                    ofstream out;
-                               ofstream out2;
-                               
-                               #ifdef USE_MPI
-                                       int pid;
-                                       MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
-                                       if (pid == 0) {  
-                               #endif
-
-                               
-                               m->openOutputFile(probFileName, out);
-                               
-                               //output mothur version
-                               out << "#" << m->getVersion() << endl;
-                               
-                               out << numKmers << endl;
-                               
-                               m->openOutputFile(probFileName2, out2);
-                               
-                               //output mothur version
-                               out2 << "#" << m->getVersion() << endl;
-                               
-                               #ifdef USE_MPI
-                                       }
-                               #endif
-
-                               
-                               //for each word
-                               for (int i = 0; i < numKmers; i++) {
-                                       if (m->control_pressed) {  break; }
-                                       
-                                       #ifdef USE_MPI
-                                               MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
-                                               if (pid == 0) {  
-                                       #endif
-
-                                       out << i << '\t';
-                                       
-                                       #ifdef USE_MPI
-                                               }
-                                       #endif
-                                       
-                                       vector<int> seqsWithWordi = database->getSequencesWithKmer(i);
-                                       
-                                       map<int, int> count;
-                                       for (int k = 0; k < genusNodes.size(); k++) {  count[genusNodes[k]] = 0;  }                     
-                                                       
-                                       //for each sequence with that word
-                                       for (int j = 0; j < seqsWithWordi.size(); j++) {
-                                               int temp = phyloTree->getIndex(names[seqsWithWordi[j]]);
-                                               count[temp]++;  //increment count of seq in this genus who have this word
-                                       }
-                                       
-                                       //probabilityInTemplate = (# of seqs with that word in template + 0.50) / (total number of seqs in template + 1);
-                                       float probabilityInTemplate = (seqsWithWordi.size() + 0.50) / (float) (names.size() + 1);
-                                       diffPair tempProb(log(probabilityInTemplate), 0.0);
-                                       WordPairDiffArr[i] = tempProb;
-                                               
-                                       int numNotZero = 0;
-                                       for (int k = 0; k < genusNodes.size(); k++) {
-                                               //probabilityInThisTaxonomy = (# of seqs with that word in this taxonomy + probabilityInTemplate) / (total number of seqs in this taxonomy + 1);
-                                               
-                                               
-                                               wordGenusProb[i][k] = log((count[genusNodes[k]] + probabilityInTemplate) / (float) (genusTotals[k] + 1));  
-                                                                       
-                                               if (count[genusNodes[k]] != 0) { 
-                                                       #ifdef USE_MPI
-                                                               int pid;
-                                                               MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-                                               
-                                                               if (pid == 0) {  
-                                                       #endif
-
-                                                       out << k << '\t' << wordGenusProb[i][k] << '\t' ; 
-                                                       
-                                                       #ifdef USE_MPI
-                                                               }
-                                                       #endif
-
-                                                       numNotZero++;  
-                                               }
-                                       }
-                                       
-                                       #ifdef USE_MPI
-                                               MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-                               
-                                               if (pid == 0) {  
-                                       #endif
-                                       
-                                       out << endl;
-                                       out2 << probabilityInTemplate << '\t' << numNotZero << '\t' << log(probabilityInTemplate) << endl;
-                                       
-                                       #ifdef USE_MPI
-                                               }
-                                       #endif
-                               }
-                               
-                               #ifdef USE_MPI
-                                       MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-                               
-                                       if (pid == 0) {  
-                               #endif
-                               
-                               out.close();
-                               out2.close();
-                               
-                               #ifdef USE_MPI
-                                       }
-                               #endif
-                               
-                               //read in new phylotree with less info. - its faster
-                               ifstream phyloTreeTest(phyloTreeName.c_str());
-                               delete phyloTree;
-                               
-                               phyloTree = new PhyloTree(phyloTreeTest, phyloTreeName);
-                
-                               //save probabilities
-                               if (rdb->save) { rdb->wordGenusProb = wordGenusProb; rdb->WordPairDiffArr = WordPairDiffArr; }
-                       }
-               }
-               
-               generateWordPairDiffArr();
-               
-               //save probabilities
-               if (rdb->save) { rdb->wordGenusProb = wordGenusProb; rdb->WordPairDiffArr = WordPairDiffArr; }
-               
-               m->mothurOut("DONE."); m->mothurOutEndLine();
-               m->mothurOut("It took " + toString(time(NULL) - start) + " seconds get probabilities. "); m->mothurOutEndLine();
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "Bayesian");
-               exit(1);
-       }
-}
-/**************************************************************************************************/
-Bayesian::~Bayesian() {
-       try {
-               
-                delete phyloTree; 
-                if (database != NULL) {  delete database; }
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "~Bayesian");
-               exit(1);
-       }
-}
-
-/**************************************************************************************************/
-string Bayesian::getTaxonomy(Sequence* seq) {
-       try {
-               string tax = "";
-               Kmer kmer(kmerSize);
-               flipped = false;
-               
-               //get words contained in query
-               //getKmerString returns a string where the index in the string is hte kmer number 
-               //and the character at that index can be converted to be the number of times that kmer was seen
-               string queryKmerString = kmer.getKmerString(seq->getUnaligned()); 
-               
-               vector<int> queryKmers;
-               for (int i = 0; i < queryKmerString.length()-1; i++) {  // the -1 is to ignore any kmer with an N in it
-                       if (queryKmerString[i] != '!') { //this kmer is in the query
-                               queryKmers.push_back(i);
-                       }
-               }
-               
-               //if user wants to test reverse compliment and its reversed use that instead
-               if (flip) {     
-                       if (isReversed(queryKmers)) { 
-                               flipped = true;
-                               seq->reverseComplement(); 
-                               queryKmerString = kmer.getKmerString(seq->getUnaligned()); 
-                               queryKmers.clear();
-                               for (int i = 0; i < queryKmerString.length()-1; i++) {  // the -1 is to ignore any kmer with an N in it
-                                       if (queryKmerString[i] != '!') { //this kmer is in the query
-                                               queryKmers.push_back(i);
-                                       }
-                               }
-                       }  
-               }
-               
-               if (queryKmers.size() == 0) {  m->mothurOut(seq->getName() + "is bad."); m->mothurOutEndLine(); simpleTax = "unknown;";  return "unknown;"; }
-               
-               
-               int index = getMostProbableTaxonomy(queryKmers);
-               
-               if (m->control_pressed) { return tax; }
-                                       
-               //bootstrap - to set confidenceScore
-               int numToSelect = queryKmers.size() / 8;
-       
-               tax = bootstrapResults(queryKmers, index, numToSelect);
-               
-               return tax;     
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "getTaxonomy");
-               exit(1);
-       }
-}
-/**************************************************************************************************/
-string Bayesian::bootstrapResults(vector<int> kmers, int tax, int numToSelect) {
-       try {
-                               
-               map<int, int> confidenceScores; 
-               
-               //initialize confidences to 0 
-               int seqIndex = tax;
-               TaxNode seq = phyloTree->get(tax);
-               confidenceScores[tax] = 0;
-               
-               while (seq.level != 0) { //while you are not at the root
-                       seqIndex = seq.parent;
-                       confidenceScores[seqIndex] = 0;
-                       seq = phyloTree->get(seq.parent);
-               }
-                               
-               map<int, int>::iterator itBoot;
-               map<int, int>::iterator itBoot2;
-               map<int, int>::iterator itConvert;
-                       
-               for (int i = 0; i < iters; i++) {
-                       if (m->control_pressed) { return "control"; }
-                       
-                       vector<int> temp;
-                       for (int j = 0; j < numToSelect; j++) {
-                               int index = int(rand() % kmers.size());
-                               
-                               //add word to temp
-                               temp.push_back(kmers[index]);
-                       }
-                       
-                       //get taxonomy
-                       int newTax = getMostProbableTaxonomy(temp);
-                       //int newTax = 1;
-                       TaxNode taxonomyTemp = phyloTree->get(newTax);
-                       
-                       //add to confidence results
-                       while (taxonomyTemp.level != 0) { //while you are not at the root
-                               itBoot2 = confidenceScores.find(newTax); //is this a classification we already have a count on
-                               
-                               if (itBoot2 != confidenceScores.end()) { //this is a classification we need a confidence for
-                                       (itBoot2->second)++;
-                               }
-                               
-                               newTax = taxonomyTemp.parent;
-                               taxonomyTemp = phyloTree->get(newTax);
-                       }
-       
-               }
-               
-               string confidenceTax = "";
-               simpleTax = "";
-               
-               int seqTaxIndex = tax;
-               TaxNode seqTax = phyloTree->get(tax);
-               
-               while (seqTax.level != 0) { //while you are not at the root
-                                       
-                               itBoot2 = confidenceScores.find(seqTaxIndex); //is this a classification we already have a count on
-                               
-                               int confidence = 0;
-                               if (itBoot2 != confidenceScores.end()) { //already in confidence scores
-                                       confidence = itBoot2->second;
-                               }
-                               
-                               if (((confidence/(float)iters) * 100) >= confidenceThreshold) {
-                                       confidenceTax = seqTax.name + "(" + toString(((confidence/(float)iters) * 100)) + ");" + confidenceTax;
-                                       simpleTax = seqTax.name + ";" + simpleTax;
-                               }
-                               
-                               seqTaxIndex = seqTax.parent;
-                               seqTax = phyloTree->get(seqTax.parent);
-               }
-               
-               if (confidenceTax == "") { confidenceTax = "unknown;"; simpleTax = "unknown;";  }
-       
-               return confidenceTax;
-               
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "bootstrapResults");
-               exit(1);
-       }
-}
-/**************************************************************************************************/
-int Bayesian::getMostProbableTaxonomy(vector<int> queryKmer) {
-       try {
-               int indexofGenus = 0;
-               
-               double maxProbability = -1000000.0;
-               //find taxonomy with highest probability that this sequence is from it
-               
-               
-//             cout << genusNodes.size() << endl;
-               
-               
-               for (int k = 0; k < genusNodes.size(); k++) {
-                       //for each taxonomy calc its probability
-                       
-                       double prob = 0.0000;
-                       for (int i = 0; i < queryKmer.size(); i++) {
-                               prob += wordGenusProb[queryKmer[i]][k];
-                       }
-                       
-//                     cout << phyloTree->get(genusNodes[k]).name << '\t' << prob << endl;
-
-                       //is this the taxonomy with the greatest probability?
-                       if (prob > maxProbability) { 
-                               indexofGenus = genusNodes[k];
-                               maxProbability = prob;
-                       }
-               }
-               
-                       
-               return indexofGenus;
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "getMostProbableTaxonomy");
-               exit(1);
-       }
-}
-//********************************************************************************************************************
-//if it is more probable that the reverse compliment kmers are in the template, then we assume the sequence is reversed.
-bool Bayesian::isReversed(vector<int>& queryKmers){
-       try{
-               bool reversed = false;
-               float prob = 0;
-               float reverseProb = 0;
-                
-        for (int i = 0; i < queryKmers.size(); i++){
-            int kmer = queryKmers[i];
-            if (kmer >= 0){
-                prob += WordPairDiffArr[kmer].prob;
-                               reverseProb += WordPairDiffArr[kmer].reverseProb;
-            }
-        }
-               
-        if (reverseProb > prob){ reversed = true; }
-       
-               return reversed;
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "isReversed");
-               exit(1);
-       }
-}
-//********************************************************************************************************************
-int Bayesian::generateWordPairDiffArr(){
-       try{
-               Kmer kmer(kmerSize);
-               for (int i = 0; i < WordPairDiffArr.size(); i++) {
-                       int reversedWord = kmer.getReverseKmerNumber(i);
-                       WordPairDiffArr[i].reverseProb = WordPairDiffArr[reversedWord].prob;
-               }
-               
-               return 0;
-       }catch(exception& e) {
-               m->errorOut(e, "Bayesian", "generateWordPairDiffArr");
-               exit(1);
-       }
-}
-/*************************************************************************************************
-map<string, int> Bayesian::parseTaxMap(string newTax) {
-       try{
-       
-               map<string, int> parsed;
-               
-               newTax = newTax.substr(0, newTax.length()-1);  //get rid of last ';'
-       
-               //parse taxonomy
-               string individual;
-               while (newTax.find_first_of(';') != -1) {
-                       individual = newTax.substr(0,newTax.find_first_of(';'));
-                       newTax = newTax.substr(newTax.find_first_of(';')+1, newTax.length());
-                       parsed[individual] = 1;
-               }
-               
-               //get last one
-               parsed[newTax] = 1;
-
-               return parsed;
-               
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "parseTax");
-               exit(1);
-       }
-}
-**************************************************************************************************/
-void Bayesian::readProbFile(ifstream& in, ifstream& inNum, string inName, string inNumName) {
-       try{
-               
-               #ifdef USE_MPI
-                       
-                       int pid, num, num2, processors;
-                       vector<unsigned long long> positions;
-                       vector<unsigned long long> positions2;
-                       
-                       MPI_Status status; 
-                       MPI_File inMPI;
-                       MPI_File inMPI2;
-                       MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-                       MPI_Comm_size(MPI_COMM_WORLD, &processors);
-                       int tag = 2001;
-
-                       char inFileName[1024];
-                       strcpy(inFileName, inNumName.c_str());
-                       
-                       char inFileName2[1024];
-                       strcpy(inFileName2, inName.c_str());
-
-                       MPI_File_open(MPI_COMM_WORLD, inFileName, MPI_MODE_RDONLY, MPI_INFO_NULL, &inMPI);  //comm, filename, mode, info, filepointer
-                       MPI_File_open(MPI_COMM_WORLD, inFileName2, MPI_MODE_RDONLY, MPI_INFO_NULL, &inMPI2);  //comm, filename, mode, info, filepointer
-
-                       if (pid == 0) {
-                               positions = m->setFilePosEachLine(inNumName, num);
-                               positions2 = m->setFilePosEachLine(inName, num2);
-                               
-                               for(int i = 1; i < processors; i++) { 
-                                       MPI_Send(&num, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
-                                       MPI_Send(&positions[0], (num+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
-                                       
-                                       MPI_Send(&num2, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
-                                       MPI_Send(&positions2[0], (num2+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
-                               }
-
-                       }else{
-                               MPI_Recv(&num, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
-                               positions.resize(num+1);
-                               MPI_Recv(&positions[0], (num+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
-                               
-                               MPI_Recv(&num2, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
-                               positions2.resize(num2+1);
-                               MPI_Recv(&positions2[0], (num2+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
-                       }
-                       
-                       //read version
-                       int length = positions2[1] - positions2[0];
-                       char* buf5 = new char[length];
-
-                       MPI_File_read_at(inMPI2, positions2[0], buf5, length, MPI_CHAR, &status);
-                       delete buf5;
-
-                       //read numKmers
-                       length = positions2[2] - positions2[1];
-                       char* buf = new char[length];
-
-                       MPI_File_read_at(inMPI2, positions2[1], buf, length, MPI_CHAR, &status);
-
-                       string tempBuf = buf;
-                       if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
-                       delete buf;
-
-                       istringstream iss (tempBuf,istringstream::in);
-                       iss >> numKmers;  
-                       
-                       //initialze probabilities
-                       wordGenusProb.resize(numKmers);
-                       
-                       for (int j = 0; j < wordGenusProb.size(); j++) {        wordGenusProb[j].resize(genusNodes.size());             }
-                       
-                       int kmer, name;  
-                       vector<int> numbers; numbers.resize(numKmers);
-                       float prob;
-                       vector<float> zeroCountProb; zeroCountProb.resize(numKmers);
-                       WordPairDiffArr.resize(numKmers);
-                       
-                       //read version
-                       length = positions[1] - positions[0];
-                       char* buf6 = new char[length];
-
-                       MPI_File_read_at(inMPI2, positions[0], buf6, length, MPI_CHAR, &status);
-                       delete buf6;
-                       
-                       //read file 
-                       for(int i=1;i<num;i++){
-                               //read next sequence
-                               length = positions[i+1] - positions[i];
-                               char* buf4 = new char[length];
-
-                               MPI_File_read_at(inMPI, positions[i], buf4, length, MPI_CHAR, &status);
-
-                               tempBuf = buf4;
-                               if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
-                               delete buf4;
-
-                               istringstream iss (tempBuf,istringstream::in);
-                               float probTemp;
-                               iss >> zeroCountProb[i] >> numbers[i] >> probTemp; 
-                               WordPairDiffArr[i].prob = probTemp;
-
-                       }
-                       
-                       MPI_File_close(&inMPI);
-                       
-                       for(int i=2;i<num2;i++){
-                               //read next sequence
-                               length = positions2[i+1] - positions2[i];
-                               char* buf4 = new char[length];
-
-                               MPI_File_read_at(inMPI2, positions2[i], buf4, length, MPI_CHAR, &status);
-
-                               tempBuf = buf4;
-                               if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
-                               delete buf4;
-
-                               istringstream iss (tempBuf,istringstream::in);
-                               
-                               iss >> kmer;
-                               
-                               //set them all to zero value
-                               for (int i = 0; i < genusNodes.size(); i++) {
-                                       wordGenusProb[kmer][i] = log(zeroCountProb[kmer] / (float) (genusTotals[i]+1));
-                               }
-                               
-                               //get probs for nonzero values
-                               for (int i = 0; i < numbers[kmer]; i++) {
-                                       iss >> name >> prob;
-                                       wordGenusProb[kmer][name] = prob;
-                               }
-                               
-                       }
-                       MPI_File_close(&inMPI2);
-                       MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
-               #else
-                       //read version
-                       string line = m->getline(in); m->gobble(in);
-                       
-                       in >> numKmers; m->gobble(in);
-                       //cout << threadID << '\t' << line << '\t' << numKmers << &in << '\t' << &inNum << '\t' << genusNodes.size() << endl;
-                       //initialze probabilities
-                       wordGenusProb.resize(numKmers);
-                       
-                       for (int j = 0; j < wordGenusProb.size(); j++) {        wordGenusProb[j].resize(genusNodes.size());             }
-                       
-                       int kmer, name, count;  count = 0;
-                       vector<int> num; num.resize(numKmers);
-                       float prob;
-                       vector<float> zeroCountProb; zeroCountProb.resize(numKmers);    
-                       WordPairDiffArr.resize(numKmers);
-                       
-                       //read version
-                       string line2 = m->getline(inNum); m->gobble(inNum);
-                       float probTemp;
-               //cout << threadID << '\t' << line2 << '\t' << this << endl;    
-                       while (inNum) {
-                               inNum >> zeroCountProb[count] >> num[count] >> probTemp; 
-                               WordPairDiffArr[count].prob = probTemp;
-                               count++;
-                               m->gobble(inNum);
-                               //cout << threadID << '\t' << count << endl;
-                       }
-                       inNum.close();
-               //cout << threadID << '\t' << "here1 " << &wordGenusProb << '\t' << &num << endl; //
-               //cout << threadID << '\t' << &genusTotals << '\t' << endl; 
-               //cout << threadID << '\t' << genusNodes.size() << endl;
-                       while(in) {
-                               in >> kmer;
-                       //cout << threadID << '\t' << kmer << endl;
-                               //set them all to zero value
-                               for (int i = 0; i < genusNodes.size(); i++) {
-                                       wordGenusProb[kmer][i] = log(zeroCountProb[kmer] / (float) (genusTotals[i]+1));
-                               }
-                       //cout << threadID << '\t' << num[kmer] << "here" << endl;      
-                               //get probs for nonzero values
-                               for (int i = 0; i < num[kmer]; i++) {
-                                       in >> name >> prob;
-                                       wordGenusProb[kmer][name] = prob;
-                               }
-                               
-                               m->gobble(in);
-                       }
-                       in.close();
-               //cout << threadID << '\t' << "here" << endl;   
-               #endif
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "readProbFile");
-               exit(1);
-       }
-}
-/**************************************************************************************************/
-bool Bayesian::checkReleaseDate(ifstream& file1, ifstream& file2, ifstream& file3, ifstream& file4) {
-       try {
-               
-               bool good = true;
-               
-               vector<string> lines;
-               lines.push_back(m->getline(file1));  
-               lines.push_back(m->getline(file2)); 
-               lines.push_back(m->getline(file3)); 
-               lines.push_back(m->getline(file4)); 
-
-               //before we added this check
-               if ((lines[0][0] != '#') || (lines[1][0] != '#') || (lines[2][0] != '#') || (lines[3][0] != '#')) {  good = false;  }
-               else {
-                       //rip off #
-                       for (int i = 0; i < lines.size(); i++) { lines[i] = lines[i].substr(1);  }
-                       
-                       //get mothurs current version
-                       string version = m->getVersion();
-                       
-                       vector<string> versionVector;
-                       m->splitAtChar(version, versionVector, '.');
-                       
-                       //check each files version
-                       for (int i = 0; i < lines.size(); i++) { 
-                               vector<string> linesVector;
-                               m->splitAtChar(lines[i], linesVector, '.');
-                       
-                               if (versionVector.size() != linesVector.size()) { good = false; break; }
-                               else {
-                                       for (int j = 0; j < versionVector.size(); j++) {
-                                               int num1, num2;
-                                               convert(versionVector[j], num1);
-                                               convert(linesVector[j], num2);
-                                               
-                                               //if mothurs version is newer than this files version, then we want to remake it
-                                               if (num1 > num2) {  good = false; break;  }
-                                       }
-                               }
-                               
-                               if (!good) { break; }
-                       }
-               }
-               
-               if (!good) {  file1.close(); file2.close(); file3.close(); file4.close();  }
-               else { file1.seekg(0);  file2.seekg(0);  file3.seekg(0);  file4.seekg(0);  }
-               
-               return good;
-       }
-       catch(exception& e) {
-               m->errorOut(e, "Bayesian", "checkReleaseDate");
-               exit(1);
-       }
-}
-/**************************************************************************************************/
-
-
-
-
-
-