rawtrainingdatamaker.cpp

   1 /*
   2  *  rawTrainingDataMaker.cpp
   3  *  Mothur
   4  *
   5  *  Created by westcott on 4/21/10.
   6  *  Copyright 2010 Schloss Lab. All rights reserved.
   7  *
   8  */
   9
  10 #include "rawtrainingdatamaker.h"
  11
  12 /**************************************************************************************************/
  13
  14 RawTrainingDataMaker::RawTrainingDataMaker(){
  15         try {
  16                 m = MothurOut::getInstance();
  17                 numNodes = 1;
  18                 numSeqs = 0;
  19                 tree.push_back(rawTaxNode("Root"));
  20                 tree[0].rank = "Root";
  21                 maxLevel = 0;
  22         }
  23         catch(exception& e) {
  24                 m->errorOut(e, "RawTrainingDataMaker", "RawTrainingDataMaker");
  25                 exit(1);
  26         }
  27 }
  28 /**************************************************************************************************/
  29
  30 RawTrainingDataMaker::RawTrainingDataMaker(string tfile){
  31         try {
  32                 m = MothurOut::getInstance();
  33                 numNodes = 1;
  34                 numSeqs = 0;
  35                 tree.push_back(rawTaxNode("Root"));
  36                 tree[0].rank = "Root";
  37                 maxLevel = 0;
  38
  39                 ifstream in;
  40                 openInputFile(tfile, in);
  41
  42                 //read in users taxonomy file and add sequences to tree
  43                 string name, tax;
  44                 while(!in.eof()){
  45                         in >> name >> tax; gobble(in);
  46
  47                         addSeqToTree(name, tax);
  48                 }
  49                 in.close();
  50
  51                 assignRank(0);
  52         }
  53         catch(exception& e) {
  54                 m->errorOut(e, "RawTrainingDataMaker", "RawTrainingDataMaker");
  55                 exit(1);
  56         }
  57 }
  58
  59 /**************************************************************************************************/
  60
  61 string RawTrainingDataMaker::getNextTaxon(string& heirarchy){
  62         try {
  63                 string currentLevel = "";
  64                 if(heirarchy != ""){
  65                         int pos = heirarchy.find_first_of(';');
  66                         currentLevel=heirarchy.substr(0,pos);
  67                         if (pos != (heirarchy.length()-1)) {  heirarchy=heirarchy.substr(pos+1);  }
  68                         else { heirarchy = ""; }
  69                 }
  70                 return currentLevel;
  71         }
  72         catch(exception& e) {
  73                 m->errorOut(e, "RawTrainingDataMaker", "getNextTaxon");
  74                 exit(1);
  75         }
  76 }
  77
  78 /**************************************************************************************************/
  79
  80 int RawTrainingDataMaker::addSeqToTree(string seqName, string seqTaxonomy){
  81         try {
  82                 numSeqs++;
  83
  84                 map<string, int>::iterator childPointer;
  85
  86                 int currentNode = 0;
  87                 string taxon;
  88
  89                 while(seqTaxonomy != ""){
  90
  91                         if (m->control_pressed) { return 0; }
  92
  93                         //somehow the parent is getting one too many accnos
  94                         //use print to reassign the taxa id
  95                         taxon = getNextTaxon(seqTaxonomy);
  96
  97                         childPointer = tree[currentNode].children.find(taxon);
  98
  99                         if(childPointer != tree[currentNode].children.end()){   //if the node already exists, move on
 100                                 currentNode = childPointer->second;
 101                         }else{                                                                                  //otherwise, create it
 102                                 tree.push_back(rawTaxNode(taxon));
 103                                 numNodes++;
 104                                 tree[currentNode].children[taxon] = numNodes-1;
 105                                 tree[numNodes-1].parent = currentNode;
 106
 107                                 currentNode = tree[currentNode].children[taxon];
 108                         }
 109                 }
 110
 111         }
 112         catch(exception& e) {
 113                 m->errorOut(e, "RawTrainingDataMaker", "addSeqToTree");
 114                 exit(1);
 115         }
 116 }
 117 /**************************************************************************************************/
 118
 119 void RawTrainingDataMaker::assignRank(int index){
 120         try {
 121                 map<string,int>::iterator it;
 122
 123                 string ranks[9] = { "Root","Domain","Kingdom","Phylum","Class","Order","Family","Genus","Species" };
 124
 125                 for(it=tree[index].children.begin();it!=tree[index].children.end();it++){
 126                         tree[it->second].level = tree[index].level + 1;
 127
 128                         if (tree[it->second].level > 8) {
 129                                 tree[it->second].rank = ("unknown" + toString(tree[it->second].level));
 130                         }else {
 131                                 tree[it->second].rank = ranks[tree[it->second].level];
 132                         }
 133
 134                         //save maxLevel for binning the unclassified seqs
 135                         if (tree[it->second].level > maxLevel) { maxLevel = tree[it->second].level; }
 136
 137                         assignRank(it->second);
 138                 }
 139         }
 140         catch(exception& e) {
 141                 m->errorOut(e, "RawTrainingDataMaker", "assignRank");
 142                 exit(1);
 143         }
 144 }
 145 /**************************************************************************************************/
 146
 147 void RawTrainingDataMaker::print(ofstream& out){
 148         try {
 149                 //string temp = tree[0].name +" " + tree[0].rank;
 150                 //sanityCheck[temp] = temp;
 151
 152                 out << "0" << "*" << tree[0].name << "*" << tree[0].parent << "*" << tree[0].level << "*" << tree[0].rank << endl;
 153                 print(0, out);
 154
 155         }
 156         catch(exception& e) {
 157                 m->errorOut(e, "RawTrainingDataMaker", "print");
 158                 exit(1);
 159         }
 160 }
 161
 162 /**************************************************************************************************/
 163
 164 void RawTrainingDataMaker::print(int i, ofstream& out){
 165         try {
 166                 map<string,int>::iterator it;
 167                 for(it=tree[i].children.begin();it!=tree[i].children.end();it++){
 168                         //string temp = tree[it->second].name + " " + tree[it->second].rank;
 169
 170                         //map<string, string>::iterator itSan;
 171                         //itSan = sanityCheck.find(temp);
 172
 173                         //if (itSan == sanityCheck.end()) {
 174                                 out << it->second << "*" << tree[it->second].name << "*" << tree[it->second].parent << "*" << tree[it->second].level << "*" << tree[it->second].rank << endl;
 175                                 //sanityCheck[temp] = temp;
 176                         //}
 177                         print(it->second, out);
 178                 }
 179         }
 180         catch(exception& e) {
 181                 m->errorOut(e, "RawTrainingDataMaker", "print");
 182                 exit(1);
 183         }
 184 }
 185 /**************************************************************************************************/
 186
 187
 188