phylotypecommand.cpp

   1 /*
   2  *  phylotypecommand.cpp
   3  *  Mothur
   4  *
   5  *  Created by westcott on 11/20/09.
   6  *  Copyright 2009 Schloss Lab. All rights reserved.
   7  *
   8  */
   9
  10 #include "phylotypecommand.h"
  11 #include "phylotree.h"
  12 #include "listvector.hpp"
  13 #include "rabundvector.hpp"
  14 #include "sabundvector.hpp"
  15
  16 //**********************************************************************************************************************
  17 vector<string> PhylotypeCommand::setParameters(){
  18         try {
  19                 CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptaxonomy);
  20                 CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
  21                 CommandParameter pcutoff("cutoff", "Number", "", "-1", "", "", "",false,false); parameters.push_back(pcutoff);
  22                 CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
  23                 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
  24                 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
  25
  26                 vector<string> myArray;
  27                 for (int i = 0; i < parameters.size(); i++) {   myArray.push_back(parameters[i].name);          }
  28                 return myArray;
  29         }
  30         catch(exception& e) {
  31                 m->errorOut(e, "PhylotypeCommand", "setParameters");
  32                 exit(1);
  33         }
  34 }
  35 //**********************************************************************************************************************
  36 string PhylotypeCommand::getHelpString(){
  37         try {
  38                 string helpString = "";
  39                 helpString += "The phylotype command reads a taxonomy file and outputs a .list, .rabund and .sabund file. \n";
  40                 helpString += "The phylotype command parameter options are taxonomy, cutoff and label. The taxonomy parameter is required.\n";
  41                 helpString += "The cutoff parameter allows you to specify the level you want to stop at.  The default is the highest level in your taxonomy file. \n";
  42                 helpString += "For example: taxonomy = Bacteria;Bacteroidetes-Chlorobi;Bacteroidetes; - cutoff=2, would truncate the taxonomy to Bacteria;Bacteroidetes-Chlorobi; \n";
  43                 helpString += "For the cutoff parameter levels count up from the root of the phylotree. This enables you to look at the grouping down to a specific resolution, say the genus level.\n";
  44                 helpString += "The label parameter allows you to specify which level you would like, and are separated by dashes.  The default all levels in your taxonomy file. \n";
  45                 helpString += "For the label parameter, levels count down from the root to keep the output similiar to mothur's other commands which report information from finer resolution to coarser resolutions.\n";
  46                 helpString += "The phylotype command should be in the following format: \n";
  47                 helpString += "phylotype(taxonomy=yourTaxonomyFile, cutoff=yourCutoff, label=yourLabels) \n";
  48                 helpString += "Eaxample: phylotype(taxonomy=amazon.taxonomy, cutoff=5, label=1-3-5).\n";
  49                 return helpString;
  50         }
  51         catch(exception& e) {
  52                 m->errorOut(e, "PhylotypeCommand", "getHelpString");
  53                 exit(1);
  54         }
  55 }
  56
  57 //**********************************************************************************************************************
  58 PhylotypeCommand::PhylotypeCommand(){
  59         try {
  60                 abort = true; calledHelp = true;
  61                 setParameters();
  62                 vector<string> tempOutNames;
  63                 outputTypes["list"] = tempOutNames;
  64                 outputTypes["sabund"] = tempOutNames;
  65                 outputTypes["rabund"] = tempOutNames;
  66         }
  67         catch(exception& e) {
  68                 m->errorOut(e, "PhylotypeCommand", "PhylotypeCommand");
  69                 exit(1);
  70         }
  71 }
  72 /**********************************************************************************************************************/
  73 PhylotypeCommand::PhylotypeCommand(string option)  {
  74         try {
  75                 abort = false; calledHelp = false;
  76
  77                 //allow user to run help
  78                 if(option == "help") { help(); abort = true; calledHelp = true; }
  79                 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
  80
  81                 else {
  82                         vector<string> myArray = setParameters();
  83
  84                         OptionParser parser(option);
  85                         map<string, string> parameters = parser.getParameters();
  86
  87                         ValidParameters validParameter;
  88                         map<string, string>::iterator it;
  89
  90                         //check to make sure all parameters are valid for command
  91                         for (it = parameters.begin(); it != parameters.end(); it++) {
  92                                 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) {  abort = true;  }
  93                         }
  94
  95                         //initialize outputTypes
  96                         vector<string> tempOutNames;
  97                         outputTypes["list"] = tempOutNames;
  98                         outputTypes["sabund"] = tempOutNames;
  99                         outputTypes["rabund"] = tempOutNames;
 100
 101                         //if the user changes the input directory command factory will send this info to us in the output parameter
 102                         string inputDir = validParameter.validFile(parameters, "inputdir", false);
 103                         if (inputDir == "not found"){   inputDir = "";          }
 104                         else {
 105                                 string path;
 106                                 it = parameters.find("taxonomy");
 107                                 //user has given a template file
 108                                 if(it != parameters.end()){
 109                                         path = m->hasPath(it->second);
 110                                         //if the user has not given a path then, add inputdir. else leave path alone.
 111                                         if (path == "") {       parameters["taxonomy"] = inputDir + it->second;         }
 112                                 }
 113
 114                                 it = parameters.find("name");
 115                                 //user has given a template file
 116                                 if(it != parameters.end()){
 117                                         path = m->hasPath(it->second);
 118                                         //if the user has not given a path then, add inputdir. else leave path alone.
 119                                         if (path == "") {       parameters["name"] = inputDir + it->second;             }
 120                                 }
 121                         }
 122
 123                         taxonomyFileName = validParameter.validFile(parameters, "taxonomy", true);
 124                         if (taxonomyFileName == "not found") {
 125                                 taxonomyFileName = m->getTaxonomyFile();
 126                                 if (taxonomyFileName != "") {  m->mothurOut("Using " + taxonomyFileName + " as input file for the taxonomy parameter."); m->mothurOutEndLine(); }
 127                                 else {
 128                                         m->mothurOut("No valid current files. taxonomy is a required parameter."); m->mothurOutEndLine();
 129                                         abort = true;
 130                                 }
 131                         }else if (taxonomyFileName == "not open") { taxonomyFileName = ""; abort = true; }
 132                         else { m->setTaxonomyFile(taxonomyFileName); }
 133
 134                         namefile = validParameter.validFile(parameters, "name", true);
 135                         if (namefile == "not open") { namefile = ""; abort = true; }
 136                         else if (namefile == "not found") { namefile = ""; }
 137                         else { readNamesFile(); m->setNameFile(namefile); }
 138
 139                         //if the user changes the output directory command factory will send this info to us in the output parameter
 140                         outputDir = validParameter.validFile(parameters, "outputdir", false);           if (outputDir == "not found"){
 141                                 outputDir = "";
 142                                 outputDir += m->hasPath(taxonomyFileName); //if user entered a file with a path then preserve it
 143                         }
 144
 145                         string temp = validParameter.validFile(parameters, "cutoff", false);
 146                         if (temp == "not found") { temp = "-1"; }
 147                         m->mothurConvert(temp, cutoff);
 148
 149                         label = validParameter.validFile(parameters, "label", false);
 150                         if (label == "not found") { label = ""; allLines = 1; }
 151                         else {
 152                                 if(label != "all") {  m->splitAtDash(label, labels);  allLines = 0;  }
 153                                 else { allLines = 1;  }
 154                         }
 155
 156                         if (namefile == "") {
 157                                 vector<string> files; files.push_back(taxonomyFileName);
 158                                 parser.getNameFile(files);
 159                         }
 160
 161                 }
 162         }
 163         catch(exception& e) {
 164                 m->errorOut(e, "PhylotypeCommand", "PhylotypeCommand");
 165                 exit(1);
 166         }
 167 }
 168 /**********************************************************************************************************************/
 169
 170 int PhylotypeCommand::execute(){
 171         try {
 172
 173                 if (abort == true) { if (calledHelp) { return 0; }  return 2;   }
 174
 175                 //reads in taxonomy file and makes all the taxonomies the same length
 176                 //by appending the last taxon to a given taxonomy as many times as needed to
 177                 //make it as long as the longest taxonomy in the file
 178                 TaxEqualizer* taxEqual = new TaxEqualizer(taxonomyFileName, cutoff, outputDir);
 179
 180                 if (m->control_pressed) { delete taxEqual; return 0; }
 181
 182                 string equalizedTaxFile = taxEqual->getEqualizedTaxFile();
 183
 184                 delete taxEqual;
 185
 186                 //build taxonomy tree from equalized file
 187                 PhyloTree* tree = new PhyloTree(equalizedTaxFile);
 188                 vector<int> leaves = tree->getGenusNodes();
 189
 190                 //store leaf nodes in current map
 191                 for (int i = 0; i < leaves.size(); i++)         {       currentNodes[leaves[i]] = leaves[i];    }
 192
 193                 bool done = false;
 194                 if (tree->get(leaves[0]).parent == -1) {  m->mothurOut("Empty Tree"); m->mothurOutEndLine();    done = true;    }
 195
 196                 if (m->control_pressed) { delete tree; return 0; }
 197
 198                 string fileroot = outputDir + m->getRootName(m->getSimpleName(taxonomyFileName));
 199
 200                 ofstream outList;
 201                 string outputListFile = fileroot + "tx.list";
 202                 m->openOutputFile(outputListFile, outList);
 203                 ofstream outSabund;
 204                 string outputSabundFile = fileroot + "tx.sabund";
 205                 m->openOutputFile(outputSabundFile, outSabund);
 206                 ofstream outRabund;
 207                 string outputRabundFile = fileroot + "tx.rabund";
 208                 m->openOutputFile(outputRabundFile, outRabund);
 209
 210                 outputNames.push_back(outputListFile); outputTypes["list"].push_back(outputListFile);
 211                 outputNames.push_back(outputSabundFile); outputTypes["sabund"].push_back(outputSabundFile);
 212                 outputNames.push_back(outputRabundFile); outputTypes["rabund"].push_back(outputRabundFile);
 213
 214                 int count = 1;
 215                 //start at leaves of tree and work towards root, processing the labels the user wants
 216                 while((!done) && ((allLines == 1) || (labels.size() != 0))) {
 217
 218                         string level = toString(count);
 219                         count++;
 220
 221                         if (m->control_pressed) {
 222                                 outRabund.close(); outSabund.close(); outList.close();
 223                                 for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]);  }
 224                                 delete tree; return 0;
 225                         }
 226
 227                         //is this a level the user want output for
 228                         if(allLines == 1 || labels.count(level) == 1){
 229
 230                                 //output level
 231                                 m->mothurOut(level); m->mothurOutEndLine();
 232
 233                                 ListVector list;
 234                                 list.setLabel(level);
 235
 236                                 //go through nodes and build listvector
 237                                 for (itCurrent = currentNodes.begin(); itCurrent != currentNodes.end(); itCurrent++) {
 238
 239                                         //get parents
 240                                         TaxNode node = tree->get(itCurrent->first);
 241                                         parentNodes[node.parent] = node.parent;
 242
 243                                         vector<string> names = node.accessions;
 244
 245                                         //make the names compatable with listvector
 246                                         string name = "";
 247                                         for (int i = 0; i < names.size(); i++) {
 248
 249                         if (names[i] != "unknown") {
 250                             if (namefile != "") {
 251                                 map<string, string>::iterator itNames = namemap.find(names[i]);  //make sure this name is in namefile
 252
 253                                 if (itNames != namemap.end()) {  name += namemap[names[i]] + ",";   } //you found it in namefile
 254                                 else { m->mothurOut(names[i] + " is not in your namefile, please correct."); m->mothurOutEndLine(); exit(1);  }
 255
 256                             }else{   name += names[i] + ",";    }
 257                         }
 258                                         }
 259                                         name = name.substr(0, name.length()-1);  //rip off extra ','
 260                                         //add bin to list vector
 261                                         if (name != "") { list.push_back(name); } //caused by unknown
 262                                 }
 263
 264                                 //print listvector
 265                                 list.print(outList);
 266                                 //print rabund
 267                                 list.getRAbundVector().print(outRabund);
 268                                 //print sabund
 269                                 list.getSAbundVector().print(outSabund);
 270
 271                                 labels.erase(level);
 272
 273                         }else {
 274
 275                                 //just get parents
 276                                 for (itCurrent = currentNodes.begin(); itCurrent != currentNodes.end(); itCurrent++) {
 277                                         int parent = tree->get(itCurrent->first).parent;
 278                                         parentNodes[parent] = parent;
 279                                 }
 280                         }
 281
 282                         //move up a level
 283                         currentNodes = parentNodes;
 284                         parentNodes.clear();
 285
 286                         //have we reached the rootnode
 287                         if (tree->get(currentNodes.begin()->first).parent == -1)  {  done = true;  }
 288                 }
 289
 290                 outList.close();
 291                 outSabund.close();
 292                 outRabund.close();
 293
 294                 delete tree;
 295
 296                 if (m->control_pressed) {
 297                         for (int i = 0; i < outputNames.size(); i++) {  m->mothurRemove(outputNames[i]);  }
 298                         return 0;
 299                 }
 300
 301                 //set list file as new current listfile
 302                 string current = "";
 303                 itTypes = outputTypes.find("list");
 304                 if (itTypes != outputTypes.end()) {
 305                         if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); }
 306                 }
 307
 308                 //set rabund file as new current rabundfile
 309                 itTypes = outputTypes.find("rabund");
 310                 if (itTypes != outputTypes.end()) {
 311                         if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRabundFile(current); }
 312                 }
 313
 314                 //set sabund file as new current sabundfile
 315                 itTypes = outputTypes.find("sabund");
 316                 if (itTypes != outputTypes.end()) {
 317                         if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSabundFile(current); }
 318                 }
 319
 320                 m->mothurOutEndLine();
 321                 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
 322                 for (int i = 0; i < outputNames.size(); i++) {  m->mothurOut(outputNames[i]); m->mothurOutEndLine();    }
 323                 m->mothurOutEndLine();
 324
 325                 return 0;
 326         }
 327
 328         catch(exception& e) {
 329                 m->errorOut(e, "PhylotypeCommand", "execute");
 330                 exit(1);
 331         }
 332 }
 333 /*****************************************************************/
 334 int PhylotypeCommand::readNamesFile() {
 335         try {
 336
 337                 ifstream in;
 338                 m->openInputFile(namefile, in);
 339
 340                 string first, second;
 341                 map<string, string>::iterator itNames;
 342
 343                 while(!in.eof()) {
 344                         in >> first >> second; m->gobble(in);
 345
 346                         itNames = namemap.find(first);
 347                         if (itNames == namemap.end()) {
 348                                 namemap[first] = second;
 349                         }else {  m->mothurOut(first + " has already been seen in namefile, disregarding names file."); m->mothurOutEndLine(); in.close(); namemap.clear(); namefile = ""; return 1; }
 350                 }
 351                 in.close();
 352
 353                 return 0;
 354         }
 355         catch(exception& e) {
 356                 m->errorOut(e, "PhylotypeCommand", "readNamesFile");
 357                 exit(1);
 358         }
 359 }
 360
 361 /**********************************************************************************************************************/