#include "unifracweightedcommand.h"
+//**********************************************************************************************************************
+vector<string> UnifracWeightedCommand::setParameters(){
+ try {
+ CommandParameter ptree("tree", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptree);
+ CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup);
+ CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
+ CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
+ CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters);
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
+ CommandParameter prandom("random", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(prandom);
+ CommandParameter pdistance("distance", "Multiple", "column-lt-square", "column", "", "", "",false,false); parameters.push_back(pdistance);
+ CommandParameter proot("root", "Boolean", "F", "", "", "", "",false,false); parameters.push_back(proot);
+ CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
+ CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
+
+ vector<string> myArray;
+ for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "setParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string UnifracWeightedCommand::getHelpString(){
+ try {
+ string helpString = "";
+ helpString += "The unifrac.weighted command parameters are tree, group, name, groups, iters, distance, processors, root and random. tree parameter is required unless you have valid current tree file.\n";
+ helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like analyzed. You must enter at least 2 valid groups.\n";
+ helpString += "The group names are separated by dashes. The iters parameter allows you to specify how many random trees you would like compared to your tree.\n";
+ helpString += "The distance parameter allows you to create a distance file from the results. The default is false.\n";
+ helpString += "The random parameter allows you to shut off the comparison to random trees. The default is false, meaning don't compare your trees with randomly generated trees.\n";
+ helpString += "The root parameter allows you to include the entire root in your calculations. The default is false, meaning stop at the root for this comparision instead of the root of the entire tree.\n";
+ helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n";
+ helpString += "The unifrac.weighted command should be in the following format: unifrac.weighted(groups=yourGroups, iters=yourIters).\n";
+ helpString += "Example unifrac.weighted(groups=A-B-C, iters=500).\n";
+ helpString += "The default value for groups is all the groups in your groupfile, and iters is 1000.\n";
+ helpString += "The unifrac.weighted command output two files: .weighted and .wsummary their descriptions are in the manual.\n";
+ helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n";
+ return helpString;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "getHelpString");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+UnifracWeightedCommand::UnifracWeightedCommand(){
+ try {
+ abort = true; calledHelp = true;
+ setParameters();
+ vector<string> tempOutNames;
+ outputTypes["weighted"] = tempOutNames;
+ outputTypes["wsummary"] = tempOutNames;
+ outputTypes["phylip"] = tempOutNames;
+ outputTypes["column"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "UnifracWeightedCommand");
+ exit(1);
+ }
+}
+
/***********************************************************/
-UnifracWeightedCommand::UnifracWeightedCommand() {
+UnifracWeightedCommand::UnifracWeightedCommand(string option) {
try {
- globaldata = GlobalData::getInstance();
-
- T = globaldata->gTree;
- tmap = globaldata->gTreemap;
- weightedFile = globaldata->getTreeFile() + ".weighted";
- openOutputFile(weightedFile, out);
- sumFile = globaldata->getTreeFile() + ".wsummary";
- openOutputFile(sumFile, outSum);
- distFile = globaldata->getTreeFile() + ".wdistrib";
- openOutputFile(distFile, outDist);
-
- //if the user has not entered specific groups to analyze then do them all
- if (globaldata->Groups.size() == 0) {
- numGroups = tmap->getNumGroups();
- }else {
- //check that groups are valid
- for (int i = 0; i < globaldata->Groups.size(); i++) {
- if (tmap->isValidGroup(globaldata->Groups[i]) != true) {
- cout << globaldata->Groups[i] << " is not a valid group, and will be disregarded." << endl;
- // erase the invalid group from globaldata->Groups
- globaldata->Groups.erase (globaldata->Groups.begin()+i);
- }
- }
+ abort = false; calledHelp = false;
- //if the user only entered invalid groups
- if (globaldata->Groups.size() == 0) {
- numGroups = tmap->getNumGroups();
- cout << "When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile." << endl;
- }else if (globaldata->Groups.size() == 1) {
- cout << "When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile." << endl;
- numGroups = tmap->getNumGroups();
- globaldata->Groups.clear();
- }else { numGroups = globaldata->Groups.size(); }
- }
+ //allow user to run help
+ if(option == "help") { help(); abort = true; calledHelp = true; }
+ else if(option == "citation") { citation(); abort = true; calledHelp = true;}
+
+ else {
+ vector<string> myArray = setParameters();
+
+ OptionParser parser(option);
+ map<string,string> parameters=parser.getParameters();
+ map<string,string>::iterator it;
+
+ ValidParameters validParameter;
- //calculate number of comparisons i.e. with groups A,B,C = AB, AC, BC = 3;
- numComp = 0;
- int n = 1;
- for (int i=1; i<numGroups; i++) {
- numComp += i;
- for (int l = n; l < numGroups; l++) {
- //set group comparison labels
- if (globaldata->Groups.size() != 0) {
- groupComb.push_back(globaldata->Groups[i-1]+globaldata->Groups[l]);
- }else {
- groupComb.push_back(tmap->namesOfGroups[i-1]+tmap->namesOfGroups[l]);
+ //check to make sure all parameters are valid for command
+ for (map<string,string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
+ if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
+ }
+
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["weighted"] = tempOutNames;
+ outputTypes["wsummary"] = tempOutNames;
+ outputTypes["phylip"] = tempOutNames;
+ outputTypes["column"] = tempOutNames;
+
+ //if the user changes the input directory command factory will send this info to us in the output parameter
+ string inputDir = validParameter.validFile(parameters, "inputdir", false);
+ if (inputDir == "not found"){ inputDir = ""; }
+ else {
+ string path;
+ it = parameters.find("tree");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["tree"] = inputDir + it->second; }
+ }
+
+ it = parameters.find("group");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = m->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; }
+ }
+
+ it = parameters.find("name");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["name"] = inputDir + it->second; }
}
}
- n++;
- }
- convert(globaldata->getIters(), iters); //how many random trees to generate
- weighted = new Weighted(tmap);
-
+ m->runParse = true;
+ m->Groups.clear();
+ m->namesOfGroups.clear();
+ m->Treenames.clear();
+ m->names.clear();
+
+ //check for required parameters
+ treefile = validParameter.validFile(parameters, "tree", true);
+ if (treefile == "not open") { abort = true; }
+ else if (treefile == "not found") { //if there is a current design file, use it
+ treefile = m->getTreeFile();
+ if (treefile != "") { m->mothurOut("Using " + treefile + " as input file for the tree parameter."); m->mothurOutEndLine(); }
+ else { m->mothurOut("You have no current tree file and the tree parameter is required."); m->mothurOutEndLine(); abort = true; }
+ }else { m->setTreeFile(treefile); }
+
+ //check for required parameters
+ groupfile = validParameter.validFile(parameters, "group", true);
+ if (groupfile == "not open") { abort = true; }
+ else if (groupfile == "not found") { groupfile = ""; }
+ else { m->setGroupFile(groupfile); }
+
+ namefile = validParameter.validFile(parameters, "name", true);
+ if (namefile == "not open") { abort = true; }
+ else if (namefile == "not found") { namefile = ""; }
+ else { m->setNameFile(namefile); }
+
+ outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
+
+
+ //check for optional parameter and set defaults
+ // ...at some point should added some additional type checking...
+ groups = validParameter.validFile(parameters, "groups", false);
+ if (groups == "not found") { groups = ""; }
+ else {
+ m->splitAtDash(groups, Groups);
+ m->Groups = Groups;
+ }
+
+ itersString = validParameter.validFile(parameters, "iters", false); if (itersString == "not found") { itersString = "1000"; }
+ convert(itersString, iters);
+
+ string temp = validParameter.validFile(parameters, "distance", false);
+ if (temp == "not found") { phylip = false; outputForm = ""; }
+ else{
+ if ((temp == "lt") || (temp == "column") || (temp == "square")) { phylip = true; outputForm = temp; }
+ else { m->mothurOut("Options for distance are: lt, square, or column. Using lt."); m->mothurOutEndLine(); phylip = true; outputForm = "lt"; }
+ }
+
+ temp = validParameter.validFile(parameters, "random", false); if (temp == "not found") { temp = "F"; }
+ random = m->isTrue(temp);
+
+ temp = validParameter.validFile(parameters, "root", false); if (temp == "not found") { temp = "F"; }
+ includeRoot = m->isTrue(temp);
+
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
+ m->setProcessors(temp);
+ convert(temp, processors);
+
+ if (!random) { iters = 0; } //turn off random calcs
+ }
+
+
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function UnifracWeightedCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the UnifracWeightedCommand class function UnifracWeightedCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ m->errorOut(e, "UnifracWeightedCommand", "UnifracWeightedCommand");
exit(1);
}
}
/***********************************************************/
int UnifracWeightedCommand::execute() {
try {
+
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
+
+ m->setTreeFile(treefile);
+
+ if (groupfile != "") {
+ //read in group map info.
+ tmap = new TreeMap(groupfile);
+ tmap->readMap();
+ }else{ //fake out by putting everyone in one group
+ Tree* tree = new Tree(treefile); delete tree; //extracts names from tree to make faked out groupmap
+ tmap = new TreeMap();
+
+ for (int i = 0; i < m->Treenames.size(); i++) { tmap->addSeq(m->Treenames[i], "Group1"); }
+ }
+
+ if (namefile != "") { readNamesFile(); }
+
+ read = new ReadNewickTree(treefile);
+ int readOk = read->read(tmap);
+
+ if (readOk != 0) { m->mothurOut("Read Terminated."); m->mothurOutEndLine(); delete tmap; delete read; return 0; }
+
+ read->AssembleTrees();
+ T = read->getTrees();
+ delete read;
+
+ //make sure all files match
+ //if you provide a namefile we will use the numNames in the namefile as long as the number of unique match the tree names size.
+ int numNamesInTree;
+ if (namefile != "") {
+ if (numUniquesInName == m->Treenames.size()) { numNamesInTree = nameMap.size(); }
+ else { numNamesInTree = m->Treenames.size(); }
+ }else { numNamesInTree = m->Treenames.size(); }
+
+
+ //output any names that are in group file but not in tree
+ if (numNamesInTree < tmap->getNumSeqs()) {
+ for (int i = 0; i < tmap->namesOfSeqs.size(); i++) {
+ //is that name in the tree?
+ int count = 0;
+ for (int j = 0; j < m->Treenames.size(); j++) {
+ if (tmap->namesOfSeqs[i] == m->Treenames[j]) { break; } //found it
+ count++;
+ }
+
+ if (m->control_pressed) {
+ delete tmap; for (int i = 0; i < T.size(); i++) { delete T[i]; }
+ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear();
+ m->Groups.clear();
+ return 0;
+ }
+
+ //then you did not find it so report it
+ if (count == m->Treenames.size()) {
+ //if it is in your namefile then don't remove
+ map<string, string>::iterator it = nameMap.find(tmap->namesOfSeqs[i]);
+
+ if (it == nameMap.end()) {
+ m->mothurOut(tmap->namesOfSeqs[i] + " is in your groupfile and not in your tree. It will be disregarded."); m->mothurOutEndLine();
+ tmap->removeSeq(tmap->namesOfSeqs[i]);
+ i--; //need this because removeSeq removes name from namesOfSeqs
+ }
+ }
+ }
+ }
+
+ sumFile = outputDir + m->getSimpleName(treefile) + ".wsummary";
+ m->openOutputFile(sumFile, outSum);
+ outputNames.push_back(sumFile); outputTypes["wsummary"].push_back(sumFile);
+
+ util = new SharedUtil();
+ string s; //to make work with setgroups
+ util->setGroups(m->Groups, tmap->namesOfGroups, s, numGroups, "weighted"); //sets the groups the user wants to analyze
+ util->getCombos(groupComb, m->Groups, numComp);
+ delete util;
+
+ weighted = new Weighted(tmap, includeRoot);
+
+ int start = time(NULL);
//get weighted for users tree
userData.resize(numComp,0); //data[0] = weightedscore AB, data[1] = weightedscore AC...
randomData.resize(numComp,0); //data[0] = weightedscore AB, data[1] = weightedscore AC...
- uscoreFreq.resize(numComp);
- validScores.resize(numComp);
- totalrscoreFreq.resize(numComp);
- uCumul.resize(numComp);
-
- //format output
- outDist.setf(ios::fixed, ios::floatfield); outDist.setf(ios::showpoint);
- outDist << "Tree#" << '\t' << "Iter" << '\t' << "Groups"<< '\t' << "WScore" << endl;
-
- //create new tree with same num nodes and leaves as users
- randT = new Tree();
-
- //get pscores for users trees
- for (int i = 0; i < T.size(); i++) {
- rscoreFreq.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC...
- rCumul.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC...
+ if (numComp < processors) { processors = numComp; }
- cout << "Processing tree " << i+1 << endl;
- userData = weighted->getValues(T[i]); //userData[0] = weightedscore
+ //get weighted scores for users trees
+ for (int i = 0; i < T.size(); i++) {
+
+ if (m->control_pressed) { delete tmap; delete weighted;
+ for (int i = 0; i < T.size(); i++) { delete T[i]; } outSum.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+
+ counter = 0;
+ rScores.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC...
+ uScores.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC...
+
+ if (random) {
+ output = new ColumnFile(outputDir + m->getSimpleName(treefile) + toString(i+1) + ".weighted", itersString);
+ outputNames.push_back(outputDir + m->getSimpleName(treefile) + toString(i+1) + ".weighted");
+ outputTypes["weighted"].push_back(outputDir + m->getSimpleName(treefile) + toString(i+1) + ".weighted");
+ }
+
+ userData = weighted->getValues(T[i], processors, outputDir); //userData[0] = weightedscore
+
+ if (m->control_pressed) { delete tmap; delete weighted;
+ for (int i = 0; i < T.size(); i++) { delete T[i]; } if (random) { delete output; } outSum.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
//save users score
for (int s=0; s<numComp; s++) {
- //update uscoreFreq
- it = uscoreFreq[s].find(userData[s]);
- if (it == uscoreFreq[s].end()) {//new score
- uscoreFreq[s][userData[s]] = 1;
- }else{ uscoreFreq[s][userData[s]]++; }
+ //add users score to vector of user scores
+ uScores[s].push_back(userData[s]);
- //add user score to valid scores
- validScores[s][userData[s]] = userData[s];
-
//save users tree score for summary file
utreeScores.push_back(userData[s]);
}
- //copy T[i]'s info.
- randT->getCopy(T[i]);
-
- //get pscores for random trees
- for (int j = 0; j < iters; j++) {
- //create a random tree with same topology as T[i], but different labels
- randT->assembleRandomUnifracTree();
- //get pscore of random tree
- randomData = weighted->getValues(randT);
+ if (random) {
+
+ //calculate number of comparisons i.e. with groups A,B,C = AB, AC, BC = 3;
+ vector< vector<string> > namesOfGroupCombos;
+ for (int a=0; a<numGroups; a++) {
+ for (int l = 0; l < a; l++) {
+ vector<string> groups; groups.push_back(m->Groups[a]); groups.push_back(m->Groups[l]);
+ namesOfGroupCombos.push_back(groups);
+ }
+ }
+
+ lines.clear();
- //save ramdoms score
- for (int p=0; p<numComp; p++) {
- //add trees weighted score random score freq
- it2 = rscoreFreq[p].find(randomData[p]);
- if (it2 != rscoreFreq[p].end()) {//already have that score
- rscoreFreq[p][randomData[p]]++;
- }else{//first time we have seen this score
- rscoreFreq[p][randomData[p]] = 1;
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors != 1){
+ int numPairs = namesOfGroupCombos.size();
+ int numPairsPerProcessor = numPairs / processors;
+
+ for (int i = 0; i < processors; i++) {
+ int startPos = i * numPairsPerProcessor;
+ if(i == processors - 1){
+ numPairsPerProcessor = numPairs - i * numPairsPerProcessor;
+ }
+ lines.push_back(linePair(startPos, numPairsPerProcessor));
+ }
}
+ #endif
+
+
+ //get scores for random trees
+ for (int j = 0; j < iters; j++) {
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors == 1){
+ driver(T[i], namesOfGroupCombos, 0, namesOfGroupCombos.size(), rScores);
+ }else{
+ createProcesses(T[i], namesOfGroupCombos, rScores);
+ }
+ #else
+ driver(T[i], namesOfGroupCombos, 0, namesOfGroupCombos.size(), rScores);
+ #endif
- //add random score to valid scores
- validScores[p][randomData[p]] = randomData[p];
+ if (m->control_pressed) { delete tmap; delete weighted;
+ for (int i = 0; i < T.size(); i++) { delete T[i]; } delete output; outSum.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
- //output info to uwdistrib file
- outDist << i+1 << '\t' << '\t'<< j+1 << '\t' << '\t' << groupComb[p] << '\t'<< randomData[p] << endl;
+ //report progress
+// m->mothurOut("Iter: " + toString(j+1)); m->mothurOutEndLine();
}
- }
+ lines.clear();
- saveRandomScores(); //save all random scores for weighted file
-
- //find the signifigance of the score for summary file
- for (int t = 0; t < numComp; t++) {
- float rcumul = 0.0000;
- for (it = validScores[t].begin(); it != validScores[t].end(); it++) {
- //make rscoreFreq map and rCumul
- it2 = rscoreFreq[t].find(it->first);
- //get percentage of random trees with that info
- if (it2 != rscoreFreq[t].end()) { rscoreFreq[t][it->first] /= iters; rcumul+= it2->second; }
- else { rscoreFreq[t][it->first] = 0.0000; } //no random trees with that score
- rCumul[t][it->first] = rcumul;
+ //find the signifigance of the score for summary file
+ for (int f = 0; f < numComp; f++) {
+ //sort random scores
+ sort(rScores[f].begin(), rScores[f].end());
+
+ //the index of the score higher than yours is returned
+ //so if you have 1000 random trees the index returned is 100
+ //then there are 900 trees with a score greater then you.
+ //giving you a signifigance of 0.900
+ int index = findIndex(userData[f], f); if (index == -1) { m->mothurOut("error in UnifracWeightedCommand"); m->mothurOutEndLine(); exit(1); } //error code
+
+ //the signifigance is the number of trees with the users score or higher
+ WScoreSig.push_back((iters-index)/(float)iters);
}
- }
+
+ //out << "Tree# " << i << endl;
+ calculateFreqsCumuls();
+ printWeightedFile();
+
+ delete output;
- //save the signifigance of the users score for printing later
- for (int f = 0; f < numComp; f++) {
- WScoreSig.push_back(rCumul[f][userData[f]]);
}
-
- //clear random data
- rscoreFreq.clear();
- rCumul.clear();
+ //clear data
+ rScores.clear();
+ uScores.clear();
+ validScores.clear();
}
- rCumul.resize(numComp);
- for (int b = 0; b < numComp; b++) {
- float ucumul = 0.0000;
- float rcumul = 0.0000;
- //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print.
- for (it = validScores[b].begin(); it != validScores[b].end(); it++) {
- it2 = uscoreFreq[b].find(it->first);
- //user data has that score
- if (it2 != uscoreFreq[b].end()) { uscoreFreq[b][it->first] /= T.size(); ucumul+= it2->second; }
- else { uscoreFreq[b][it->first] = 0.0000; } //no user trees with that score
- //make uCumul map
- uCumul[b][it->first] = ucumul;
-
- //make rscoreFreq map and rCumul
- it2 = totalrscoreFreq[b].find(it->first);
- //get percentage of random trees with that info
- if (it2 != totalrscoreFreq[b].end()) { totalrscoreFreq[b][it->first] /= (iters * T.size()); rcumul+= it2->second; }
- else { totalrscoreFreq[b][it->first] = 0.0000; } //no random trees with that score
- rCumul[b][it->first] = rcumul;
- }
- }
- printWeightedFile();
+ if (m->control_pressed) { delete tmap; delete weighted;
+ for (int i = 0; i < T.size(); i++) { delete T[i]; } outSum.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+
printWSummaryFile();
- //reset randomTree parameter to 0
- globaldata->setRandomTree("0");
+ if (phylip) { createPhylipFile(); }
+
//clear out users groups
- globaldata->Groups.clear();
+ m->Groups.clear();
+ delete tmap; delete weighted;
+ for (int i = 0; i < T.size(); i++) { delete T[i]; }
- delete randT;
+
+ if (m->control_pressed) {
+ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
+ return 0;
+ }
+
+ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to run unifrac.weighted."); m->mothurOutEndLine();
+
+ //set phylip file as new current phylipfile
+ string current = "";
+ itTypes = outputTypes.find("phylip");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setPhylipFile(current); }
+ }
+
+ //set column file as new current columnfile
+ itTypes = outputTypes.find("column");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setColumnFile(current); }
+ }
+
+ m->mothurOutEndLine();
+ m->mothurOut("Output File Names: "); m->mothurOutEndLine();
+ for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
+ m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the UnifracWeightedCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ m->errorOut(e, "UnifracWeightedCommand", "execute");
exit(1);
}
}
-/***********************************************************/
-void UnifracWeightedCommand::printWeightedFile() {
+/**************************************************************************************************/
+
+int UnifracWeightedCommand::createProcesses(Tree* t, vector< vector<string> > namesOfGroupCombos, vector< vector<double> >& scores) {
try {
- //column headers
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ int process = 1;
+ vector<int> processIDS;
+
+ EstOutput results;
- out << "Group" << '\t' << "Score" << '\t' << "UserFreq" << '\t' << "UserCumul" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl;
+ //loop through and create all the processes you want
+ while (process != processors) {
+ int pid = fork();
+
+ if (pid > 0) {
+ processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
+ process++;
+ }else if (pid == 0){
+ driver(t, namesOfGroupCombos, lines[process].start, lines[process].num, scores);
+
+ //pass numSeqs to parent
+ ofstream out;
+ string tempFile = outputDir + toString(getpid()) + ".weightedcommand.results.temp";
+ m->openOutputFile(tempFile, out);
+ for (int i = lines[process].start; i < (lines[process].start + lines[process].num); i++) { out << scores[i][(scores[i].size()-1)] << '\t'; } out << endl;
+ out.close();
- //format output
- out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
+ exit(0);
+ }else {
+ m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
+ for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
+ exit(0);
+ }
+ }
- //for each group
- for (int e = 0; e < numComp; e++) {
- //print each line in that group
- for (it = validScores[e].begin(); it != validScores[e].end(); it++) {
- out << setprecision(6) << groupComb[e] << '\t' << it->first << '\t' << '\t' << uscoreFreq[e][it->first] << '\t' << uCumul[e][it->first] << '\t' << totalrscoreFreq[e][it->first] << '\t' << rCumul[e][it->first] << endl;
- }
+ driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, scores);
+
+ //force parent to wait until all the processes are done
+ for (int i=0;i<(processors-1);i++) {
+ int temp = processIDS[i];
+ wait(&temp);
}
- out.close();
+ //get data created by processes
+ for (int i=0;i<(processors-1);i++) {
+
+ ifstream in;
+ string s = outputDir + toString(processIDS[i]) + ".weightedcommand.results.temp";
+ m->openInputFile(s, in);
+
+ double tempScore;
+ for (int j = lines[(i+1)].start; j < (lines[(i+1)].start + lines[(i+1)].num); j++) { in >> tempScore; scores[j].push_back(tempScore); }
+ in.close();
+ remove(s.c_str());
+ }
+
+ return 0;
+#endif
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "createProcesses");
+ exit(1);
+ }
+}
+
+/**************************************************************************************************/
+int UnifracWeightedCommand::driver(Tree* t, vector< vector<string> > namesOfGroupCombos, int start, int num, vector< vector<double> >& scores) {
+ try {
+ Tree* randT = new Tree(tmap);
+
+ for (int h = start; h < (start+num); h++) {
+
+ if (m->control_pressed) { return 0; }
+
+ //initialize weighted score
+ string groupA = namesOfGroupCombos[h][0];
+ string groupB = namesOfGroupCombos[h][1];
+
+ //copy T[i]'s info.
+ randT->getCopy(t);
+
+ //create a random tree with same topology as T[i], but different labels
+ randT->assembleRandomUnifracTree(groupA, groupB);
+
+ if (m->control_pressed) { delete randT; return 0; }
+
+ //get wscore of random tree
+ EstOutput randomData = weighted->getValues(randT, groupA, groupB);
+ if (m->control_pressed) { delete randT; return 0; }
+
+ //save scores
+ scores[h].push_back(randomData[0]);
+ }
+
+ delete randT;
+
+ return 0;
+
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function printWeightedFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ m->errorOut(e, "UnifracWeightedCommand", "driver");
exit(1);
}
- catch(...) {
- cout << "An unknown error has occurred in the UnifracWeightedCommand class function printWeightedFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+}
+/***********************************************************/
+void UnifracWeightedCommand::printWeightedFile() {
+ try {
+ vector<double> data;
+ vector<string> tags;
+ tags.push_back("Score"); tags.push_back("RandFreq"); tags.push_back("RandCumul");
+
+ for(int a = 0; a < numComp; a++) {
+ output->initFile(groupComb[a], tags);
+ //print each line
+ for (map<float,float>::iterator it = validScores.begin(); it != validScores.end(); it++) {
+ data.push_back(it->first); data.push_back(rScoreFreq[a][it->first]); data.push_back(rCumul[a][it->first]);
+ output->output(data);
+ data.clear();
+ }
+ output->resetFile();
+ }
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "printWeightedFile");
exit(1);
}
}
void UnifracWeightedCommand::printWSummaryFile() {
try {
//column headers
- outSum << "Tree#" << '\t' << "Groups" << '\t' << '\t' << "WScore" << '\t' << '\t' << "WSig" << endl;
+ outSum << "Tree#" << '\t' << "Groups" << '\t' << "WScore" << '\t' << "WSig" << endl;
+ m->mothurOut("Tree#\tGroups\tWScore\tWSig"); m->mothurOutEndLine();
//format output
outSum.setf(ios::fixed, ios::floatfield); outSum.setf(ios::showpoint);
int count = 0;
for (int i = 0; i < T.size(); i++) {
for (int j = 0; j < numComp; j++) {
- outSum << setprecision(6) << i+1 << '\t' << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << WScoreSig[count] << endl;
+ if (random) {
+ if (WScoreSig[count] > (1/(float)iters)) {
+ outSum << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(itersString.length()) << WScoreSig[count] << endl;
+ cout << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(itersString.length()) << WScoreSig[count] << endl;
+ m->mothurOutJustToLog(toString(i+1) +"\t" + groupComb[j] +"\t" + toString(utreeScores[count]) +"\t" + toString(WScoreSig[count]) + "\n");
+ }else{
+ outSum << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(itersString.length()) << "<" << (1/float(iters)) << endl;
+ cout << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << setprecision(itersString.length()) << "<" << (1/float(iters)) << endl;
+ m->mothurOutJustToLog(toString(i+1) +"\t" + groupComb[j] +"\t" + toString(utreeScores[count]) +"\t<" + toString((1/float(iters))) + "\n");
+ }
+ }else{
+ outSum << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << "0.00" << endl;
+ cout << setprecision(6) << i+1 << '\t' << groupComb[j] << '\t' << utreeScores[count] << '\t' << "0.00" << endl;
+ m->mothurOutJustToLog(toString(i+1) +"\t" + groupComb[j] +"\t" + toString(utreeScores[count]) +"\t0.00\n");
+ }
count++;
}
}
outSum.close();
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function printWeightedFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ m->errorOut(e, "UnifracWeightedCommand", "printWSummaryFile");
+ exit(1);
+ }
+}
+/***********************************************************/
+void UnifracWeightedCommand::createPhylipFile() {
+ try {
+ int count = 0;
+ //for each tree
+ for (int i = 0; i < T.size(); i++) {
+
+ string phylipFileName;
+ if ((outputForm == "lt") || (outputForm == "square")) {
+ phylipFileName = outputDir + m->getSimpleName(treefile) + toString(i+1) + ".weighted.phylip.dist";
+ outputNames.push_back(phylipFileName); outputTypes["phylip"].push_back(phylipFileName);
+ }else { //column
+ phylipFileName = outputDir + m->getSimpleName(treefile) + toString(i+1) + ".weighted.column.dist";
+ outputNames.push_back(phylipFileName); outputTypes["column"].push_back(phylipFileName);
+ }
+
+ ofstream out;
+ m->openOutputFile(phylipFileName, out);
+
+ if ((outputForm == "lt") || (outputForm == "square")) {
+ //output numSeqs
+ out << m->Groups.size() << endl;
+ }
+
+ //make matrix with scores in it
+ vector< vector<float> > dists; dists.resize(m->Groups.size());
+ for (int i = 0; i < m->Groups.size(); i++) {
+ dists[i].resize(m->Groups.size(), 0.0);
+ }
+
+ //flip it so you can print it
+ for (int r=0; r<m->Groups.size(); r++) {
+ for (int l = 0; l < r; l++) {
+ dists[r][l] = utreeScores[count];
+ dists[l][r] = utreeScores[count];
+ count++;
+ }
+ }
+
+ //output to file
+ for (int r=0; r<m->Groups.size(); r++) {
+ //output name
+ string name = m->Groups[r];
+ if (name.length() < 10) { //pad with spaces to make compatible
+ while (name.length() < 10) { name += " "; }
+ }
+
+ if (outputForm == "lt") {
+ out << name << '\t';
+
+ //output distances
+ for (int l = 0; l < r; l++) { out << dists[r][l] << '\t'; }
+ out << endl;
+ }else if (outputForm == "square") {
+ out << name << '\t';
+
+ //output distances
+ for (int l = 0; l < m->Groups.size(); l++) { out << dists[r][l] << '\t'; }
+ out << endl;
+ }else{
+ //output distances
+ for (int l = 0; l < r; l++) {
+ string otherName = m->Groups[l];
+ if (otherName.length() < 10) { //pad with spaces to make compatible
+ while (otherName.length() < 10) { otherName += " "; }
+ }
+
+ out << name << '\t' << otherName << dists[r][l] << endl;
+ }
+ }
+ }
+ out.close();
+ }
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "createPhylipFile");
exit(1);
}
- catch(...) {
- cout << "An unknown error has occurred in the UnifracWeightedCommand class function printWeightedFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+}
+/***********************************************************/
+int UnifracWeightedCommand::findIndex(float score, int index) {
+ try{
+ for (int i = 0; i < rScores[index].size(); i++) {
+ if (rScores[index][i] >= score) { return i; }
+ }
+ return rScores[index].size();
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "findIndex");
exit(1);
}
}
/***********************************************************/
-void UnifracWeightedCommand::saveRandomScores() {
+
+void UnifracWeightedCommand::calculateFreqsCumuls() {
try {
- for (int e = 0; e < numComp; e++) {
- //update total map with new random scores
- for (it = rscoreFreq[e].begin(); it != rscoreFreq[e].end(); it++) {
- //does this score already exist in the total map
- it2 = totalrscoreFreq[e].find(it->first);
- //if yes then add them
- if (it2 != totalrscoreFreq[e].end()) {
- totalrscoreFreq[e][it->first] += rscoreFreq[e][it->first];
- }else{ //its a new score
- totalrscoreFreq[e][it->first] = rscoreFreq[e][it->first];
+ //clear out old tree values
+ rScoreFreq.clear();
+ rScoreFreq.resize(numComp);
+ rCumul.clear();
+ rCumul.resize(numComp);
+ validScores.clear();
+
+ //calculate frequency
+ for (int f = 0; f < numComp; f++) {
+ for (int i = 0; i < rScores[f].size(); i++) { //looks like 0,0,1,1,1,2,4,7... you want to make a map that say rScoreFreq[0] = 2, rScoreFreq[1] = 3...
+ validScores[rScores[f][i]] = rScores[f][i];
+ map<float,float>::iterator it = rScoreFreq[f].find(rScores[f][i]);
+ if (it != rScoreFreq[f].end()) {
+ rScoreFreq[f][rScores[f][i]]++;
+ }else{
+ rScoreFreq[f][rScores[f][i]] = 1;
}
}
}
+
+ //calculate rcumul
+ for(int a = 0; a < numComp; a++) {
+ float rcumul = 1.0000;
+ //this loop fills the cumulative maps and put 0.0000 in the score freq map to make it easier to print.
+ for (map<float,float>::iterator it = validScores.begin(); it != validScores.end(); it++) {
+ //make rscoreFreq map and rCumul
+ map<float,float>::iterator it2 = rScoreFreq[a].find(it->first);
+ rCumul[a][it->first] = rcumul;
+ //get percentage of random trees with that info
+ if (it2 != rScoreFreq[a].end()) { rScoreFreq[a][it->first] /= iters; rcumul-= it2->second; }
+ else { rScoreFreq[a][it->first] = 0.0000; } //no random trees with that score
+ }
+ }
+
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function saveRandomScores. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ m->errorOut(e, "UnifracWeightedCommand", "calculateFreqsCums");
exit(1);
}
- catch(...) {
- cout << "An unknown error has occurred in the UnifracWeightedCommand class function saveRandomScores. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+}
+/*****************************************************************/
+int UnifracWeightedCommand::readNamesFile() {
+ try {
+ m->names.clear();
+ numUniquesInName = 0;
+
+ ifstream in;
+ m->openInputFile(namefile, in);
+
+ string first, second;
+ map<string, string>::iterator itNames;
+
+ while(!in.eof()) {
+ in >> first >> second; m->gobble(in);
+
+ numUniquesInName++;
+
+ itNames = m->names.find(first);
+ if (itNames == m->names.end()) {
+ m->names[first] = second;
+
+ //we need a list of names in your namefile to use above when removing extra seqs above so we don't remove them
+ vector<string> dupNames;
+ m->splitAtComma(second, dupNames);
+
+ for (int i = 0; i < dupNames.size(); i++) {
+ nameMap[dupNames[i]] = dupNames[i];
+ if ((groupfile == "") && (i != 0)) { tmap->addSeq(dupNames[i], "Group1"); }
+ }
+ }else { m->mothurOut(first + " has already been seen in namefile, disregarding names file."); m->mothurOutEndLine(); in.close(); m->names.clear(); namefile = ""; return 1; }
+ }
+ in.close();
+
+ return 0;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "readNamesFile");
exit(1);
}
}
-
/***********************************************************/
+
+
+
+
+
projects / mothur.git / blobdiff