#include "unifracweightedcommand.h"
/***********************************************************/
-UnifracWeightedCommand::UnifracWeightedCommand() {
+UnifracWeightedCommand::UnifracWeightedCommand(string option) {
try {
globaldata = GlobalData::getInstance();
+ abort = false;
+ Groups.clear();
- 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);
+ //allow user to run help
+ if(option == "help") { help(); abort = true; }
- //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);
- }
+ else {
+ //valid paramters for this command
+ string Array[] = {"groups","iters"};
+ vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+
+ OptionParser parser(option);
+ map<string,string> parameters=parser.getParameters();
+
+ ValidParameters validParameter;
+
+ //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(); }
- }
+ if (globaldata->gTree.size() == 0) {//no trees were read
+ cout << "You must execute the read.tree command, before you may execute the unifrac.weighted command." << endl; abort = true; }
+
+ //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 {
+ splitAtDash(groups, Groups);
+ globaldata->Groups = Groups;
+ }
+
+ itersString = validParameter.validFile(parameters, "iters", false); if (itersString == "not found") { itersString = "1000"; }
+ convert(itersString, iters);
- //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 (abort == false) {
+ T = globaldata->gTree;
+ tmap = globaldata->gTreemap;
+ sumFile = globaldata->getTreeFile() + ".wsummary";
+ openOutputFile(sumFile, outSum);
+
+ 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);
+
}
- n++;
}
-
- convert(globaldata->getIters(), iters); //how many random trees to generate
- 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";
exit(1);
}
}
+//**********************************************************************************************************************
+
+void UnifracWeightedCommand::help(){
+ try {
+ cout << "The unifrac.weighted command can only be executed after a successful read.tree command." << "\n";
+ cout << "The unifrac.weighted command parameters are groups and iters. No parameters are required." << "\n";
+ cout << "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";
+ cout << "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";
+ cout << "The unifrac.weighted command should be in the following format: unifrac.weighted(groups=yourGroups, iters=yourIters)." << "\n";
+ cout << "Example unifrac.weighted(groups=A-B-C, iters=500)." << "\n";
+ cout << "The default value for groups is all the groups in your groupfile, and iters is 1000." << "\n";
+ cout << "The unifrac.weighted command output two files: .weighted and .wsummary their descriptions are in the manual." << "\n";
+ cout << "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups)." << "\n" << "\n";
+ }
+ catch(exception& e) {
+ cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+ catch(...) {
+ cout << "An unknown error has occurred in the UnifracWeightedCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+}
+
/***********************************************************/
int UnifracWeightedCommand::execute() {
try {
+
+ if (abort == true) { return 0; }
+
+ Progress* reading;
+ reading = new Progress("Comparing to random:", iters);
//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;
+ counter = 0;
+ rScores.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC...
+ uScores.resize(numComp); //data[0] = weightedscore AB, data[1] = weightedscore AC...
+
+ output = new ColumnFile(globaldata->getTreeFile() + toString(i+1) + ".weighted", itersString);
+
userData = weighted->getValues(T[i]); //userData[0] = weightedscore
//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
+ //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;
+ int count = 0;
+ for (int r=0; r<numGroups; r++) {
+ for (int l = r+1; l < numGroups; l++) {
+ //copy T[i]'s info.
+ randT->getCopy(T[i]);
+
+ //create a random tree with same topology as T[i], but different labels
+ randT->assembleRandomUnifracTree(globaldata->Groups[r], globaldata->Groups[l]);
+ //get wscore of random tree
+ randomData = weighted->getValues(randT, globaldata->Groups[r], globaldata->Groups[l]);
+
+ //save scores
+ rScores[count].push_back(randomData[0]);
+ count++;
}
-
- //add random score to valid scores
- validScores[p][randomData[p]] = randomData[p];
-
- //output info to uwdistrib file
- outDist << i+1 << '\t' << '\t'<< j+1 << '\t' << '\t' << groupComb[p] << '\t'<< randomData[p] << endl;
}
+
+ //update progress bar
+ reading->update(j);
+
}
-
- saveRandomScores(); //save all random scores for weighted file
-
+
+ //removeValidScoresDuplicates();
//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;
- }
- }
-
- //save the signifigance of the users score for printing later
for (int f = 0; f < numComp; f++) {
- WScoreSig.push_back(rCumul[f][userData[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) { cout << "error in UnifracWeightedCommand" << endl; 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();
- //clear random data
- rscoreFreq.clear();
- rCumul.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;
+ delete output;
- //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;
- }
+ //clear data
+ rScores.clear();
+ uScores.clear();
+ validScores.clear();
}
- printWeightedFile();
+ //finish progress bar
+ reading->finish();
+ delete reading;
+
printWSummaryFile();
- //reset randomTree parameter to 0
- globaldata->setRandomTree("0");
//clear out users groups
globaldata->Groups.clear();
/***********************************************************/
void UnifracWeightedCommand::printWeightedFile() {
try {
- //column headers
-
- out << "Group" << '\t' << "Score" << '\t' << "UserFreq" << '\t' << "UserCumul" << '\t' << "RandFreq" << '\t' << "RandCumul" << endl;
-
- //format output
- out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
+ vector<double> data;
+ vector<string> tags;
+ tags.push_back("Score"); tags.push_back("RandFreq"); tags.push_back("RandCumul");
- //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;
+ 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();
}
-
- out.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";
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;
+ cout << "Tree#" << '\t' << "Groups" << '\t' << "WScore" << '\t' << "WSig" << endl;
//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 (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;
+ }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;
+ }
count++;
}
}
}
/***********************************************************/
-void UnifracWeightedCommand::saveRandomScores() {
+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) {
+ cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function findIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+ catch(...) {
+ cout << "An unknown error has occurred in the UnifracWeightedCommand class function findIndex. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+}
+
+/***********************************************************/
+
+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";
+ cout << "Standard Error: " << e.what() << " has occurred in the UnifracWeightedCommand class Function calculateFreqsCums. 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";
+ cout << "An unknown error has occurred in the UnifracWeightedCommand class function calculateFreqsCums. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
exit(1);
}
+
}
/***********************************************************/
+
+
+
+
+
projects / mothur.git / blobdiff