#include "unifracweightedcommand.h"
/***********************************************************/
-UnifracWeightedCommand::UnifracWeightedCommand() {
+UnifracWeightedCommand::UnifracWeightedCommand(string option) {
try {
globaldata = GlobalData::getInstance();
+ abort = false;
+ Groups.clear();
+
+ //allow user to run help
+ if(option == "help") { help(); abort = true; }
- 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);
+ else {
+ //valid paramters for this command
+ string Array[] = {"groups","iters","distance","random","processors","outputdir","inputdir"};
+ vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+
+ OptionParser parser(option);
+ map<string,string> parameters=parser.getParameters();
+
+ ValidParameters validParameter;
- //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);
- }
+ //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; }
}
- //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(); }
- }
-
- //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]);
- }
+ if (globaldata->gTree.size() == 0) {//no trees were read
+ m->mothurOut("You must execute the read.tree command, before you may execute the unifrac.weighted command."); m->mothurOutEndLine(); abort = true; }
+
+ //if the user changes the output directory command factory will send this info to us in the output parameter
+ outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
+ outputDir = "";
+ outputDir += m->hasPath(globaldata->inputFileName); //if user entered a file with a path then preserve it
}
- n++;
- }
+
+ //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);
+ globaldata->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") { temp = "false"; }
+ phylip = m->isTrue(temp);
+
+ temp = validParameter.validFile(parameters, "random", false); if (temp == "not found") { temp = "F"; }
+ random = m->isTrue(temp);
- convert(globaldata->getIters(), iters); //how many random trees to generate
- weighted = new Weighted(tmap);
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
+ convert(temp, processors);
+
+ if (!random) { iters = 0; } //turn off random calcs
+
+ if (abort == false) {
+ T = globaldata->gTree;
+ tmap = globaldata->gTreemap;
+ sumFile = outputDir + m->getSimpleName(globaldata->getTreeFile()) + ".wsummary";
+ m->openOutputFile(sumFile, outSum);
+ outputNames.push_back(sumFile);
+
+ util = new SharedUtil();
+ string s; //to make work with setgroups
+ util->setGroups(globaldata->Groups, tmap->namesOfGroups, s, numGroups, "weighted"); //sets the groups the user wants to analyze
+ util->getCombos(groupComb, globaldata->Groups, numComp);
+
+ weighted = new Weighted(tmap);
+
+ }
+ }
+
+
}
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";
+ m->errorOut(e, "UnifracWeightedCommand", "UnifracWeightedCommand");
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";
+}
+//**********************************************************************************************************************
+
+void UnifracWeightedCommand::help(){
+ try {
+ m->mothurOut("The unifrac.weighted command can only be executed after a successful read.tree command.\n");
+ m->mothurOut("The unifrac.weighted command parameters are groups, iters, distance and random. No parameters are required.\n");
+ m->mothurOut("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");
+ m->mothurOut("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");
+ m->mothurOut("The distance parameter allows you to create a distance file from the results. The default is false.\n");
+ m->mothurOut("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");
+ m->mothurOut("The unifrac.weighted command should be in the following format: unifrac.weighted(groups=yourGroups, iters=yourIters).\n");
+ m->mothurOut("Example unifrac.weighted(groups=A-B-C, iters=500).\n");
+ m->mothurOut("The default value for groups is all the groups in your groupfile, and iters is 1000.\n");
+ m->mothurOut("The unifrac.weighted command output two files: .weighted and .wsummary their descriptions are in the manual.\n");
+ m->mothurOut("Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n\n");
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "help");
exit(1);
}
}
+
/***********************************************************/
int UnifracWeightedCommand::execute() {
try {
+
+ if (abort == true) { return 0; }
+
+ 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
+ //get weighted scores 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...
-
- cout << "Processing tree " << i+1 << endl;
- userData = weighted->getValues(T[i]); //userData[0] = weightedscore
+
+ if (m->control_pressed) { delete randT; 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(globaldata->getTreeFile()) + toString(i+1) + ".weighted", itersString);
+ outputNames.push_back(outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted");
+ }
+
+ userData = weighted->getValues(T[i], processors, outputDir); //userData[0] = weightedscore
+
+ if (m->control_pressed) {
+ delete randT;
+ 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]);
+ if (random) {
+ vector<double> sums = weighted->getBranchLengthSums(T[i]);
- //get scores 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);
-
- //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;
+ //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 = a+1; l < numGroups; l++) {
+ vector<string> groups; groups.push_back(globaldata->Groups[a]); groups.push_back(globaldata->Groups[l]);
+ namesOfGroupCombos.push_back(groups);
}
+ }
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors != 1){
+ int numPairs = namesOfGroupCombos.size();
+ int numPairsPerProcessor = numPairs / processors;
- //add random score to valid scores
- validScores[p][randomData[p]] = randomData[p];
+ for (int i = 0; i < processors; i++) {
+ int startPos = i * numPairsPerProcessor;
+ if(i == processors - 1){
+ numPairsPerProcessor = numPairs - i * numPairsPerProcessor;
+ }
+ lines.push_back(new linePair(startPos, numPairsPerProcessor));
+ }
+ }
+ #endif
+
+
+ //get scores for random trees
+ for (int j = 0; j < iters; j++) {
- //output info to uwdistrib file
- outDist << i+1 << '\t' << '\t'<< j+1 << '\t' << '\t' << groupComb[p] << '\t'<< randomData[p] << endl;
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors == 1){
+ driver(T[i], randT, namesOfGroupCombos, 0, namesOfGroupCombos.size(), sums, rScores);
+ }else{
+ createProcesses(T[i], randT, namesOfGroupCombos, sums, rScores);
+ }
+ #else
+ driver(T[i], randT, namesOfGroupCombos, 0, namesOfGroupCombos.size(), sums, rScores);
+ #endif
+
+ if (m->control_pressed) { delete output; outSum.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+
+ //report progress
+ m->mothurOut("Iter: " + toString(j+1)); m->mothurOutEndLine();
}
- }
-
- 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 = 1.0000;
- for (it = validScores[t].begin(); it != validScores[t].end(); it++) {
- //make rscoreFreq map and rCumul
- it2 = rscoreFreq[t].find(it->first);
- rCumul[t][it->first] = rcumul;
- //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
+
+ for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
+
+ //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 = 1.0000;
- 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 (it = validScores[b].begin(); it != validScores[b].end(); it++) {
- it2 = uscoreFreq[b].find(it->first);
- //make uCumul map
- uCumul[b][it->first] = ucumul;
- //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 rscoreFreq map and rCumul
- it2 = totalrscoreFreq[b].find(it->first);
- rCumul[b][it->first] = rcumul;
- //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
- }
- }
- printWeightedFile();
+ if (m->control_pressed) { delete randT; 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();
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();
+
+ 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, Tree* randT, vector< vector<string> > namesOfGroupCombos, vector<double>& sums, vector< vector<double> >& scores) {
try {
- //column headers
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ int process = 1;
+ int num = 0;
+ 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, randT, namesOfGroupCombos, lines[process]->start, lines[process]->num, sums, 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("unable to spawn the necessary processes."); m->mothurOutEndLine(); 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, randT, namesOfGroupCombos, lines[0]->start, lines[0]->num, sums, 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);
+ }
+
+ //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 i = lines[process]->start; i < (lines[process]->start + lines[process]->num); i++) { in >> tempScore; scores[i].push_back(tempScore); }
+ in.close();
+ remove(s.c_str());
}
- out.close();
+ return 0;
+#endif
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "createProcesses");
+ exit(1);
+ }
+}
+
+/**************************************************************************************************/
+int UnifracWeightedCommand::driver(Tree* t, Tree* randT, vector< vector<string> > namesOfGroupCombos, int start, int num, vector<double>& sums, vector< vector<double> >& scores) {
+ try {
+
+ for (int h = start; h < (start+num); h++) {
+ cout << "doing " << h << endl;
+ 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, sums);
+
+ if (m->control_pressed) { delete randT; return 0; }
+
+ //save scores
+ scores[h].push_back(randomData[0]);
+ }
+ 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])); m->mothurOutEndLine();
+ }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)))); m->mothurOutEndLine();
+ }
+ }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"); m->mothurOutEndLine();
+ }
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);
}
- 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::createPhylipFile() {
+ try {
+ int count = 0;
+ //for each tree
+ for (int i = 0; i < T.size(); i++) {
+
+ string phylipFileName = outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted.dist";
+ outputNames.push_back(phylipFileName);
+ ofstream out;
+ m->openOutputFile(phylipFileName, out);
+
+ //output numSeqs
+ out << globaldata->Groups.size() << endl;
+
+ //make matrix with scores in it
+ vector< vector<float> > dists; dists.resize(globaldata->Groups.size());
+ for (int i = 0; i < globaldata->Groups.size(); i++) {
+ dists[i].resize(globaldata->Groups.size(), 0.0);
+ }
+
+ //flip it so you can print it
+ for (int r=0; r<globaldata->Groups.size(); r++) {
+ for (int l = r+1; l < globaldata->Groups.size(); l++) {
+ dists[r][l] = utreeScores[count];
+ dists[l][r] = utreeScores[count];
+ count++;
+ }
+ }
+
+ //output to file
+ for (int r=0; r<globaldata->Groups.size(); r++) {
+ //output name
+ string name = globaldata->Groups[r];
+ if (name.length() < 10) { //pad with spaces to make compatible
+ while (name.length() < 10) { name += " "; }
+ }
+ out << name << '\t';
+
+ //output distances
+ for (int l = 0; l < r; l++) { out << dists[r][l] << '\t'; }
+ out << endl;
+ }
+ out.close();
+ }
+ }
+ catch(exception& e) {
+ m->errorOut(e, "UnifracWeightedCommand", "createPhylipFile");
+ exit(1);
+ }
+}
+/***********************************************************/
+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";
- 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";
+ m->errorOut(e, "UnifracWeightedCommand", "calculateFreqsCums");
exit(1);
}
}
/***********************************************************/
+
+
+
+
+
projects / mothur.git / blobdiff