#include "unifracunweightedcommand.h"
/***********************************************************/
-UnifracUnweightedCommand::UnifracUnweightedCommand() {
+UnifracUnweightedCommand::UnifracUnweightedCommand(string option) {
try {
globaldata = GlobalData::getInstance();
+ abort = false;
+ Groups.clear();
+
+ //allow user to run help
+ if(option == "help") { help(); abort = true; }
+
+ 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 (globaldata->gTree.size() == 0) {//no trees were read
+ cout << "You must execute the read.tree command, before you may execute the unifrac.unweighted 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);
+
+
+ if (abort == false) {
+ T = globaldata->gTree;
+ tmap = globaldata->gTreemap;
+ sumFile = globaldata->getTreeFile() + ".uwsummary";
+ openOutputFile(sumFile, outSum);
+
+ util = new SharedUtil();
+ util->setGroups(globaldata->Groups, tmap->namesOfGroups, allGroups, numGroups, "unweighted"); //sets the groups the user wants to analyze
+ util->getCombos(groupComb, globaldata->Groups, numComp);
+
+ if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); }
+
+ unweighted = new Unweighted(tmap);
+
+ }
+
+ }
- T = globaldata->gTree;
- tmap = globaldata->gTreemap;
- sumFile = globaldata->getTreeFile() + ".uwsummary";
- openOutputFile(sumFile, outSum);
-
- setGroups(); //sets users groups to analyze
- convert(globaldata->getIters(), iters); //how many random trees to generate
- unweighted = new Unweighted(tmap);
-
}
catch(exception& e) {
cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function UnifracUnweightedCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
exit(1);
}
}
+
+//**********************************************************************************************************************
+
+void UnifracUnweightedCommand::help(){
+ try {
+ cout << "The unifrac.unweighted command can only be executed after a successful read.tree command." << "\n";
+ cout << "The unifrac.unweighted 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 1 valid group." << "\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.unweighted command should be in the following format: unifrac.unweighted(groups=yourGroups, iters=yourIters)." << "\n";
+ cout << "Example unifrac.unweighted(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.unweighted command output two files: .unweighted and .uwsummary 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 UnifracUnweightedCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+ catch(...) {
+ cout << "An unknown error has occurred in the UnifracUnweightedCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+}
+
+
/***********************************************************/
int UnifracUnweightedCommand::execute() {
try {
-
+
+ if (abort == true) { return 0; }
+
userData.resize(numComp,0); //data[0] = unweightedscore
randomData.resize(numComp,0); //data[0] = unweightedscore
//create new tree with same num nodes and leaves as users
-
+
+ outSum << "Tree#" << '\t' << "Groups" << '\t' << "UWScore" <<'\t' << "UWSig" << endl;
+ cout << "Tree#" << '\t' << "Groups" << '\t' << "UWScore" << '\t' << "UWSig" << endl;
+
//get pscores for users trees
for (int i = 0; i < T.size(); i++) {
- unweightedFile = globaldata->getTreeFile() + toString(i+1) + ".unweighted";
- unweightedFileout = globaldata->getTreeFile() + "temp." + toString(i+1) + ".unweighted";
- //column headers
- outSum << "Tree# " << i+1 << endl;
- outSum << "Comb" << '\t' << "UWScore" << '\t' << '\t' << "UWSig" << endl;
-
+ counter = 0;
+
+ output = new ColumnFile(globaldata->getTreeFile() + toString(i+1) + ".unweighted", itersString);
+
//get unweighted for users tree
rscoreFreq.resize(numComp);
rCumul.resize(numComp);
- validScores.resize(numComp);
utreeScores.resize(numComp);
UWScoreSig.resize(numComp);
- cout << "Processing tree " << i+1 << endl;
userData = unweighted->getValues(T[i]); //userData[0] = unweightedscore
//output scores for each combination
for(int k = 0; k < numComp; k++) {
- //add users score to valid scores
- validScores[k][userData[k]] = userData[k];
-
//saves users score
utreeScores[k].push_back(userData[k]);
+
}
//get unweighted scores for random trees
randomData = unweighted->getValues(T[i], "", "");
for(int k = 0; k < numComp; k++) {
-//cout << "iter " << j << " comp " << k << " = " << randomData[k] << endl;
//add trees unweighted score to map of scores
- it2 = rscoreFreq[k].find(randomData[k]);
- if (it2 != rscoreFreq[k].end()) {//already have that score
+ map<float,float>::iterator it = rscoreFreq[k].find(randomData[k]);
+ if (it != rscoreFreq[k].end()) {//already have that score
rscoreFreq[k][randomData[k]]++;
}else{//first time we have seen this score
rscoreFreq[k][randomData[k]] = 1;
}
//add randoms score to validscores
- validScores[k][randomData[k]] = randomData[k];
+ validScores[randomData[k]] = randomData[k];
}
+
}
- 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 (it = validScores[a].begin(); it != validScores[a].end(); it++) {
- //make rscoreFreq map and rCumul
- 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
+ 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
+ }
+ UWScoreSig[a].push_back(rCumul[a][userData[a]]);
}
- UWScoreSig[a].push_back(rCumul[a][userData[a]]);
- }
- printUnweightedFile();
- printUWSummaryFile();
- rscoreFreq.clear();
- rCumul.clear();
- validScores.clear();
- utreeScores.clear();
- UWScoreSig.clear();
- }
+
+ printUnweightedFile();
+ printUWSummaryFile(i);
+
+ delete output;
+ rscoreFreq.clear();
+ rCumul.clear();
+ validScores.clear();
+ utreeScores.clear();
+ UWScoreSig.clear();
+ }
+
//reset groups parameter
- globaldata->Groups.clear(); globaldata->setGroups("");
+ globaldata->Groups.clear();
outSum.close();
return 0;
void UnifracUnweightedCommand::printUnweightedFile() {
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++) {
- initFile(groupComb[a]);
+ output->initFile(groupComb[a], tags);
//print each line
- for (it = validScores[a].begin(); it != validScores[a].end(); it++) {
+ 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(data);
+ output->output(data);
data.clear();
}
- resetFile();
+ output->resetFile();
}
-
- out.close();
- inFile.close();
- remove(unweightedFileout.c_str());
-
}
catch(exception& e) {
cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function printUnweightedFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
}
/***********************************************************/
-void UnifracUnweightedCommand::printUWSummaryFile() {
+void UnifracUnweightedCommand::printUWSummaryFile(int i) {
try {
//format output
outSum.setf(ios::fixed, ios::floatfield); outSum.setf(ios::showpoint);
-
+
//print each line
- for(int a = 0; a < numComp; a++) {
- outSum << setprecision(6) << groupComb[a] << '\t' << '\t' << utreeScores[a][0] << '\t' << UWScoreSig[a][0] << endl;
- cout << setprecision(6) << groupComb[a] << '\t' << '\t' << utreeScores[a][0] << '\t' << UWScoreSig[a][0] << endl;
- }
-
- }
- catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function printUWSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the UnifracUnweightedCommand class function printUWSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
-}
-/***********************************************************/
-void UnifracUnweightedCommand::setGroups() {
- try {
- string allGroups = "";
- numGroups = 0;
- //if the user has not entered specific groups to analyze then do them all
- if (globaldata->Groups.size() != 0) {
- if (globaldata->Groups[0] != "all") {
- //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);
- }
- }
+ for(int a = 0; a < numComp; a++) {
+ outSum << i+1 << '\t';
+ cout << i+1 << '\t';
- //if the user only entered invalid groups
- if (globaldata->Groups.size() == 0) {
- cout << "When using the groups parameter you must have at least 1 valid group. I will run the command using all the groups in your groupfile." << endl;
- for (int i = 0; i < tmap->namesOfGroups.size(); i++) {
- globaldata->Groups.push_back(tmap->namesOfGroups[i]);
- numGroups++;
- allGroups += tmap->namesOfGroups[i];
- }
- }else {
- for (int i = 0; i < globaldata->Groups.size(); i++) {
- allGroups += globaldata->Groups[i];
- numGroups++;
- }
- }
- }else{//user has enter "all" and wants the default groups
- for (int i = 0; i < tmap->namesOfGroups.size(); i++) {
- globaldata->Groups.push_back(tmap->namesOfGroups[i]);
- numGroups++;
- allGroups += tmap->namesOfGroups[i];
- }
- globaldata->setGroups("");
- }
- }else {
- for (int i = 0; i < tmap->namesOfGroups.size(); i++) {
- allGroups += tmap->namesOfGroups[i];
+ if (UWScoreSig[a][0] > (1/(float)iters)) {
+ outSum << setprecision(6) << groupComb[a] << '\t' << utreeScores[a][0] << '\t' << setprecision(itersString.length()) << UWScoreSig[a][0] << endl;
+ cout << setprecision(6) << groupComb[a] << '\t' << utreeScores[a][0] << '\t' << setprecision(itersString.length()) << UWScoreSig[a][0] << endl;
+ }else {
+ outSum << setprecision(6) << groupComb[a] << '\t' << utreeScores[a][0] << '\t' << setprecision(itersString.length()) << "<" << (1/float(iters)) << endl;
+ cout << setprecision(6) << groupComb[a] << '\t' << utreeScores[a][0] << '\t' << setprecision(itersString.length()) << "<" << (1/float(iters)) << endl;
}
- numGroups = 1;
}
- //calculate number of comparsions
- numComp = 0;
- for (int r=0; r<numGroups; r++) {
- for (int l = r+1; l < numGroups; l++) {
- groupComb.push_back(globaldata->Groups[r]+globaldata->Groups[l]);
- numComp++;
- }
- }
-
- //ABC
- if (numComp != 1) {
- groupComb.push_back(allGroups);
- numComp++;
- }
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the UnifracUnweightedCommand class function setGroups. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
-
-}
-/*****************************************************************/
-
-void UnifracUnweightedCommand::initFile(string label){
- try {
- if(counter != 0){
- openOutputFile(unweightedFileout, out);
- openInputFile(unweightedFile, inFile);
-
- string inputBuffer;
- getline(inFile, inputBuffer);
-
- out << inputBuffer << '\t' << label + "Score" << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl;
- }else{
- openOutputFile(unweightedFileout, out);
- out << label + "Score" << '\t' << label + "RandFreq" << '\t' << label + "RandCumul" << endl;
- }
-
- out.setf(ios::fixed, ios::floatfield);
- out.setf(ios::showpoint);
- }
- catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function initFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function printUWSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
exit(1);
}
catch(...) {
- cout << "An unknown error has occurred in the UnifracUnweightedCommand class function initFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ cout << "An unknown error has occurred in the UnifracUnweightedCommand class function printUWSummaryFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
exit(1);
}
}
-/***********************************************************************/
-
-void UnifracUnweightedCommand::output(vector<double> data){
- try {
- if(counter != 0){
- string inputBuffer;
- getline(inFile, inputBuffer);
-
- out << inputBuffer << setprecision(6) << '\t' << data[0] << '\t' << data[1] << '\t' << data[2] << endl;
- }
- else{
- out << setprecision(6) << data[0] << '\t' << data[1] << '\t' << data[2] << endl;
- }
-
- }
- catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function output. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the UnifracUnweightedCommand class function output. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
-};
-
-/***********************************************************************/
+/***********************************************************/
-void UnifracUnweightedCommand::resetFile(){
- try {
- if(counter != 0){
- out.close();
- inFile.close();
- }
- else{
- out.close();
- }
- counter = 1;
-
- remove(unweightedFile.c_str());
- rename(unweightedFileout.c_str(), unweightedFile.c_str());
- }
- catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the UnifracUnweightedCommand class Function resetFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the UnifracUnweightedCommand class function resetFile. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
-}
projects / mothur.git / blobdiff