#include "phylodiversitycommand.h"
+//**********************************************************************************************************************
+vector<string> PhyloDiversityCommand::setParameters(){
+ try {
+
+ CommandParameter ptree("tree", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(ptree);
+ CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,true); 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 pfreq("freq", "Number", "", "100", "", "", "",false,false); parameters.push_back(pfreq);
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
+ CommandParameter prarefy("rarefy", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(prarefy);
+ CommandParameter psummary("summary", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(psummary);
+ CommandParameter pcollect("collect", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pcollect);
+ CommandParameter pscale("scale", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pscale);
+ 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, "PhyloDiversityCommand", "setParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string PhyloDiversityCommand::getHelpString(){
+ try {
+ string helpString = "";
+ helpString += "The phylo.diversity command parameters are tree, group, name, groups, iters, freq, processors, scale, rarefy, collect and summary. tree and group are required, unless you have valid current files.\n";
+ helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like analyzed. The group names are separated by dashes. By default all groups are used.\n";
+ helpString += "The iters parameter allows you to specify the number of randomizations to preform, by default iters=1000, if you set rarefy to true.\n";
+ helpString += "The freq parameter is used indicate when to output your data, by default it is set to 100. But you can set it to a percentage of the number of sequence. For example freq=0.10, means 10%. \n";
+ helpString += "The scale parameter is used indicate that you want your output scaled to the number of sequences sampled, default = false. \n";
+ helpString += "The rarefy parameter allows you to create a rarefaction curve. The default is false.\n";
+ helpString += "The collect parameter allows you to create a collectors curve. The default is false.\n";
+ helpString += "The summary parameter allows you to create a .summary file. The default is true.\n";
+ helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n";
+ helpString += "The phylo.diversity command should be in the following format: phylo.diversity(groups=yourGroups, rarefy=yourRarefy, iters=yourIters).\n";
+ helpString += "Example phylo.diversity(groups=A-B-C, rarefy=T, iters=500).\n";
+ helpString += "The phylo.diversity command output two files: .phylo.diversity and if rarefy=T, .rarefaction.\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, "PhyloDiversityCommand", "getHelpString");
+ exit(1);
+ }
+}
+
+
+//**********************************************************************************************************************
+PhyloDiversityCommand::PhyloDiversityCommand(){
+ try {
+ abort = true; calledHelp = true;
+ setParameters();
+ vector<string> tempOutNames;
+ outputTypes["phylodiv"] = tempOutNames;
+ outputTypes["rarefy"] = tempOutNames;
+ outputTypes["summary"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloDiversityCommand", "PhyloDiversityCommand");
+ exit(1);
+ }
+}
//**********************************************************************************************************************
PhyloDiversityCommand::PhyloDiversityCommand(string option) {
try {
- globaldata = GlobalData::getInstance();
- abort = false;
+ abort = false; calledHelp = false;
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true; }
else {
- //valid paramters for this command
- string Array[] = {"freq","rarefy","iters","groups","summary","collect","scale","outputdir","inputdir"};
- vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ vector<string> myArray = setParameters();;
OptionParser parser(option);
map<string,string> parameters = parser.getParameters();
+ map<string,string>::iterator it;
ValidParameters validParameter;
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { 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 = hasPath(globaldata->getTreeFile()); }
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["phylodiv"] = tempOutNames;
+ outputTypes["rarefy"] = tempOutNames;
+ outputTypes["summary"] = 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; }
+ }
+ }
+
+ 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; }
+ }
+
+ //check for required parameters
+ groupfile = validParameter.validFile(parameters, "group", true);
+ if (groupfile == "not open") { abort = true; }
+ else if (groupfile == "not found") {
+ //if there is a current design file, use it
+ groupfile = m->getGroupFile();
+ if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter."); m->mothurOutEndLine(); }
+ else { m->mothurOut("You have no current group file and the group parameter is required."); m->mothurOutEndLine(); abort = true; }
+ }
+
+ namefile = validParameter.validFile(parameters, "name", true);
+ if (namefile == "not open") { abort = true; }
+ else if (namefile == "not found") { namefile = ""; }
+
+ outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = m->hasPath(treefile); }
- if (globaldata->gTree.size() == 0) {//no trees were read
- m->mothurOut("You must execute the read.tree command, before you may execute the phylo.diversity command."); m->mothurOutEndLine(); abort = true; }
-
string temp;
temp = validParameter.validFile(parameters, "freq", false); if (temp == "not found") { temp = "100"; }
convert(temp, freq);
convert(temp, iters);
temp = validParameter.validFile(parameters, "rarefy", false); if (temp == "not found") { temp = "F"; }
- rarefy = isTrue(temp);
+ rarefy = m->isTrue(temp);
if (!rarefy) { iters = 1; }
temp = validParameter.validFile(parameters, "summary", false); if (temp == "not found") { temp = "T"; }
- summary = isTrue(temp);
+ summary = m->isTrue(temp);
temp = validParameter.validFile(parameters, "scale", false); if (temp == "not found") { temp = "F"; }
- scale = isTrue(temp);
+ scale = m->isTrue(temp);
temp = validParameter.validFile(parameters, "collect", false); if (temp == "not found") { temp = "F"; }
- collect = isTrue(temp);
+ collect = m->isTrue(temp);
+
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
+ m->setProcessors(temp);
+ convert(temp, processors);
groups = validParameter.validFile(parameters, "groups", false);
- if (groups == "not found") { groups = ""; Groups = globaldata->gTreemap->namesOfGroups; globaldata->Groups = Groups; }
+ if (groups == "not found") { groups = ""; }
else {
- splitAtDash(groups, Groups);
- globaldata->Groups = Groups;
+ m->splitAtDash(groups, Groups);
+ m->Groups = Groups;
}
if ((!collect) && (!rarefy) && (!summary)) { m->mothurOut("No outputs selected. You must set either collect, rarefy or summary to true, summary=T by default."); m->mothurOutEndLine(); abort=true; }
}
//**********************************************************************************************************************
-void PhyloDiversityCommand::help(){
- try {
- m->mothurOut("The phylo.diversity command can only be executed after a successful read.tree command.\n");
- m->mothurOut("The phylo.diversity command parameters are groups, iters, freq, scale, rarefy, collect and summary. 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. The group names are separated by dashes. By default all groups are used.\n");
- m->mothurOut("The iters parameter allows you to specify the number of randomizations to preform, by default iters=1000, if you set rarefy to true.\n");
- m->mothurOut("The freq parameter is used indicate when to output your data, by default it is set to 100. But you can set it to a percentage of the number of sequence. For example freq=0.10, means 10%. \n");
- m->mothurOut("The scale parameter is used indicate that you want your ouptut scaled to the number of sequences sampled, default = false. \n");
- m->mothurOut("The rarefy parameter allows you to create a rarefaction curve. The default is false.\n");
- m->mothurOut("The collect parameter allows you to create a collectors curve. The default is false.\n");
- m->mothurOut("The summary parameter allows you to create a .summary file. The default is true.\n");
- m->mothurOut("The phylo.diversity command should be in the following format: phylo.diversity(groups=yourGroups, rarefy=yourRarefy, iters=yourIters).\n");
- m->mothurOut("Example phylo.diversity(groups=A-B-C, rarefy=T, iters=500).\n");
- m->mothurOut("The phylo.diversity command output two files: .phylo.diversity and if rarefy=T, .rarefaction.\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, "PhyloDiversityCommand", "help");
- exit(1);
- }
-}
-
-//**********************************************************************************************************************
-
-PhyloDiversityCommand::~PhyloDiversityCommand(){}
-
-//**********************************************************************************************************************
-
int PhyloDiversityCommand::execute(){
try {
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
+
+ m->setTreeFile(treefile);
+
+ //read in group map info.
+ tmap = new TreeMap(groupfile);
+ tmap->readMap();
+
+ 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();
+ vector<Tree*> trees = 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 < trees.size(); i++) { delete trees[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
+ }
+ }
+ }
+ }
+
+ SharedUtil* util = new SharedUtil();
+ util->setGroups(m->Groups, tmap->namesOfGroups, "treegroup"); //sets the groups the user wants to analyze
+ delete util;
//incase the user had some mismatches between the tree and group files we don't want group xxx to be analyzed
- for (int i = 0; i < globaldata->Groups.size(); i++) { if (globaldata->Groups[i] == "xxx") { globaldata->Groups.erase(globaldata->Groups.begin()+i); break; } }
+ for (int i = 0; i < m->Groups.size(); i++) { if (m->Groups[i] == "xxx") { m->Groups.erase(m->Groups.begin()+i); break; } }
vector<string> outputNames;
-
- vector<Tree*> trees = globaldata->gTree;
//for each of the users trees
for(int i = 0; i < trees.size(); i++) {
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+ if (m->control_pressed) { delete tmap; for (int j = 0; j < trees.size(); j++) { delete trees[j]; } for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; }
ofstream outSum, outRare, outCollect;
- string outSumFile = outputDir + getRootName(getSimpleName(globaldata->getTreeFile())) + toString(i+1) + ".phylodiv.summary";
- string outRareFile = outputDir + getRootName(getSimpleName(globaldata->getTreeFile())) + toString(i+1) + ".phylodiv.rarefaction";
- string outCollectFile = outputDir + getRootName(getSimpleName(globaldata->getTreeFile())) + toString(i+1) + ".phylodiv";
+ string outSumFile = outputDir + m->getRootName(m->getSimpleName(treefile)) + toString(i+1) + ".phylodiv.summary";
+ string outRareFile = outputDir + m->getRootName(m->getSimpleName(treefile)) + toString(i+1) + ".phylodiv.rarefaction";
+ string outCollectFile = outputDir + m->getRootName(m->getSimpleName(treefile)) + toString(i+1) + ".phylodiv";
- if (summary) { openOutputFile(outSumFile, outSum); outputNames.push_back(outSumFile); }
- if (rarefy) { openOutputFile(outRareFile, outRare); outputNames.push_back(outRareFile); }
- if (collect) { openOutputFile(outCollectFile, outCollect); outputNames.push_back(outCollectFile); }
+ if (summary) { m->openOutputFile(outSumFile, outSum); outputNames.push_back(outSumFile); outputTypes["summary"].push_back(outSumFile); }
+ if (rarefy) { m->openOutputFile(outRareFile, outRare); outputNames.push_back(outRareFile); outputTypes["rarefy"].push_back(outRareFile); }
+ if (collect) { m->openOutputFile(outCollectFile, outCollect); outputNames.push_back(outCollectFile); outputTypes["phylodiv"].push_back(outCollectFile); }
int numLeafNodes = trees[i]->getNumLeaves();
//create a vector containing indexes of leaf nodes, randomize it, select nodes to send to calculator
vector<int> randomLeaf;
for (int j = 0; j < numLeafNodes; j++) {
- if (inUsersGroups(trees[i]->tree[j].getGroup(), globaldata->Groups) == true) { //is this a node from the group the user selected.
+ if (m->inUsersGroups(trees[i]->tree[j].getGroup(), m->Groups) == true) { //is this a node from the group the user selected.
randomLeaf.push_back(j);
}
}
//find largest group total
int largestGroup = 0;
- for (int j = 0; j < globaldata->Groups.size(); j++) {
- if (globaldata->gTreemap->seqsPerGroup[globaldata->Groups[j]] > largestGroup) { largestGroup = globaldata->gTreemap->seqsPerGroup[globaldata->Groups[j]]; }
+ for (int j = 0; j < m->Groups.size(); j++) {
+ if (tmap->seqsPerGroup[m->Groups[j]] > largestGroup) { largestGroup = tmap->seqsPerGroup[m->Groups[j]]; }
//initialize diversity
- diversity[globaldata->Groups[j]].resize(globaldata->gTreemap->seqsPerGroup[globaldata->Groups[j]]+1, 0.0); //numSampled
+ diversity[m->Groups[j]].resize(tmap->seqsPerGroup[m->Groups[j]]+1, 0.0); //numSampled
//groupA 0.0 0.0
//initialize sumDiversity
- sumDiversity[globaldata->Groups[j]].resize(globaldata->gTreemap->seqsPerGroup[globaldata->Groups[j]]+1, 0.0);
+ sumDiversity[m->Groups[j]].resize(tmap->seqsPerGroup[m->Groups[j]]+1, 0.0);
}
//convert freq percentage to number
if(largestGroup % increment != 0){ numSampledList.insert(largestGroup); }
//add other groups ending points
- for (int j = 0; j < globaldata->Groups.size(); j++) {
- if (numSampledList.count(diversity[globaldata->Groups[j]].size()-1) == 0) { numSampledList.insert(diversity[globaldata->Groups[j]].size()-1); }
+ for (int j = 0; j < m->Groups.size(); j++) {
+ if (numSampledList.count(diversity[m->Groups[j]].size()-1) == 0) { numSampledList.insert(diversity[m->Groups[j]].size()-1); }
}
+
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ if(processors == 1){
+ driver(trees[i], diversity, sumDiversity, iters, increment, randomLeaf, numSampledList, outCollect, outSum, true);
+ }else{
+ if (rarefy) {
+ vector<int> procIters;
+
+ int numItersPerProcessor = iters / processors;
+
+ //divide iters between processes
+ for (int h = 0; h < processors; h++) {
+ if(h == processors - 1){
+ numItersPerProcessor = iters - h * numItersPerProcessor;
+ }
+ procIters.push_back(numItersPerProcessor);
+ }
+
+ createProcesses(procIters, trees[i], diversity, sumDiversity, iters, increment, randomLeaf, numSampledList, outCollect, outSum);
+
+ }else{ //no need to paralellize if you dont want to rarefy
+ driver(trees[i], diversity, sumDiversity, iters, increment, randomLeaf, numSampledList, outCollect, outSum, true);
+ }
+ }
+
+ #else
+ driver(trees[i], diversity, sumDiversity, iters, increment, randomLeaf, numSampledList, outCollect, outSum, true);
+ #endif
- for (int l = 0; l < iters; l++) {
+ if (rarefy) { printData(numSampledList, sumDiversity, outRare, iters); }
+ }
+
+
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+
+ 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) {
+ m->errorOut(e, "PhyloDiversityCommand", "execute");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+int PhyloDiversityCommand::createProcesses(vector<int>& procIters, Tree* t, map< string, vector<float> >& div, map<string, vector<float> >& sumDiv, int numIters, int increment, vector<int>& randomLeaf, set<int>& numSampledList, ofstream& outCollect, ofstream& outSum){
+ try {
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ int process = 1;
+
+ vector<int> processIDS;
+ map< string, vector<float> >::iterator itSum;
+
+ //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, div, sumDiv, procIters[process], increment, randomLeaf, numSampledList, outCollect, outSum, false);
+
+ string outTemp = outputDir + toString(getpid()) + ".sumDiv.temp";
+ ofstream out;
+ m->openOutputFile(outTemp, out);
+
+ //output the sumDIversity
+ for (itSum = sumDiv.begin(); itSum != sumDiv.end(); itSum++) {
+ out << itSum->first << '\t' << (itSum->second).size() << '\t';
+ for (int k = 0; k < (itSum->second).size(); k++) {
+ out << (itSum->second)[k] << '\t';
+ }
+ out << endl;
+ }
+
+ out.close();
+
+ 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);
+ }
+ }
+
+ driver(t, div, sumDiv, procIters[0], increment, randomLeaf, numSampledList, outCollect, outSum, true);
+
+ //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++) {
+
+ //input the sumDIversity
+ string inTemp = outputDir + toString(processIDS[i]) + ".sumDiv.temp";
+ ifstream in;
+ m->openInputFile(inTemp, in);
+
+ //output the sumDIversity
+ for (int j = 0; j < sumDiv.size(); j++) {
+ string group = "";
+ int size = 0;
+
+ in >> group >> size; m->gobble(in);
+
+ for (int k = 0; k < size; k++) {
+ float tempVal;
+ in >> tempVal;
+
+ sumDiv[group][k] += tempVal;
+ }
+ m->gobble(in);
+ }
+
+ in.close();
+ remove(inTemp.c_str());
+ }
+
+#endif
+
+ return 0;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloDiversityCommand", "createProcesses");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+int PhyloDiversityCommand::driver(Tree* t, map< string, vector<float> >& div, map<string, vector<float> >& sumDiv, int numIters, int increment, vector<int>& randomLeaf, set<int>& numSampledList, ofstream& outCollect, ofstream& outSum, bool doSumCollect){
+ try {
+ int numLeafNodes = randomLeaf.size();
+
+ for (int l = 0; l < numIters; l++) {
random_shuffle(randomLeaf.begin(), randomLeaf.end());
//initialize counts
map<string, int> counts;
- for (int j = 0; j < globaldata->Groups.size(); j++) { counts[globaldata->Groups[j]] = 0; }
+ map< string, set<int> > countedBranch;
+ for (int j = 0; j < m->Groups.size(); j++) { counts[m->Groups[j]] = 0; countedBranch[m->Groups[j]].insert(-2); } //add dummy index to initialize countedBranch sets
for(int k = 0; k < numLeafNodes; k++){
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+ if (m->control_pressed) { return 0; }
//calc branch length of randomLeaf k
- float br = calcBranchLength(trees[i], randomLeaf[k]);
+ vector<float> br = calcBranchLength(t, randomLeaf[k], countedBranch);
//for each group in the groups update the total branch length accounting for the names file
- vector<string> groups = trees[i]->tree[randomLeaf[k]].getGroup();
+ vector<string> groups = t->tree[randomLeaf[k]].getGroup();
+
for (int j = 0; j < groups.size(); j++) {
int numSeqsInGroupJ = 0;
map<string, int>::iterator it;
- it = trees[i]->tree[randomLeaf[k]].pcount.find(groups[j]);
- if (it != trees[i]->tree[randomLeaf[k]].pcount.end()) { //this leaf node contains seqs from group j
+ it = t->tree[randomLeaf[k]].pcount.find(groups[j]);
+ if (it != t->tree[randomLeaf[k]].pcount.end()) { //this leaf node contains seqs from group j
numSeqsInGroupJ = it->second;
}
-
- for (int s = (counts[groups[j]]+1); s <= (counts[groups[j]]+numSeqsInGroupJ); s++) {
- diversity[groups[j]][s] = diversity[groups[j]][s-1] + ((float) numSeqsInGroupJ * br);
+
+ if (numSeqsInGroupJ != 0) { div[groups[j]][(counts[groups[j]]+1)] = div[groups[j]][counts[groups[j]]] + br[j]; }
+
+ for (int s = (counts[groups[j]]+2); s <= (counts[groups[j]]+numSeqsInGroupJ); s++) {
+ div[groups[j]][s] = div[groups[j]][s-1]; //update counts, but don't add in redundant branch lengths
}
counts[groups[j]] += numSeqsInGroupJ;
}
if (rarefy) {
//add this diversity to the sum
- for (int j = 0; j < globaldata->Groups.size(); j++) {
- for (int g = 0; g < diversity[globaldata->Groups[j]].size(); g++) {
- sumDiversity[globaldata->Groups[j]][g] += diversity[globaldata->Groups[j]][g];
+ for (int j = 0; j < m->Groups.size(); j++) {
+ for (int g = 0; g < div[m->Groups[j]].size(); g++) {
+ sumDiv[m->Groups[j]][g] += div[m->Groups[j]][g];
}
}
}
- if ((collect) && (l == 0)) { printData(numSampledList, diversity, outCollect, 1); }
- if ((summary) && (l == 0)) { printSumData(diversity, outSum, 1); }
+ if ((collect) && (l == 0) && doSumCollect) { printData(numSampledList, div, outCollect, 1); }
+ if ((summary) && (l == 0) && doSumCollect) { printSumData(div, outSum, 1); }
}
- if (rarefy) { printData(numSampledList, sumDiversity, outRare, iters); }
- }
-
-
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+ return 0;
- 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) {
- m->errorOut(e, "PhyloDiversityCommand", "execute");
+ m->errorOut(e, "PhyloDiversityCommand", "driver");
exit(1);
}
}
+
//**********************************************************************************************************************
void PhyloDiversityCommand::printSumData(map< string, vector<float> >& div, ofstream& out, int numIters){
try {
- out << "numSampled\t";
- for (int i = 0; i < globaldata->Groups.size(); i++) { out << globaldata->Groups[i] << '\t'; }
- out << endl;
+ out << "Groups\tnumSampled\tphyloDiversity" << endl;
out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
-
- set<int> num;
- //find end points to output
- for (map<string, vector<float> >::iterator itEnds = div.begin(); itEnds != div.end(); itEnds++) { num.insert(itEnds->second.size()-1); }
-
- for (set<int>::iterator it = num.begin(); it != num.end(); it++) {
- int numSampled = *it;
-
- out << numSampled << '\t';
- for (int j = 0; j < globaldata->Groups.size(); j++) {
- if (numSampled < div[globaldata->Groups[j]].size()) {
- float score;
- if (scale) { score = (div[globaldata->Groups[j]][numSampled] / (float)numIters) / (float)numSampled; }
- else { score = div[globaldata->Groups[j]][numSampled] / (float)numIters; }
-
- out << setprecision(4) << score << '\t';
- }else { out << "NA" << '\t'; }
- }
- out << endl;
- }
+ for (int j = 0; j < m->Groups.size(); j++) {
+ int numSampled = (div[m->Groups[j]].size()-1);
+ out << m->Groups[j] << '\t' << numSampled << '\t';
+
+ float score;
+ if (scale) { score = (div[m->Groups[j]][numSampled] / (float)numIters) / (float)numSampled; }
+ else { score = div[m->Groups[j]][numSampled] / (float)numIters; }
+
+ out << setprecision(4) << score << endl;
+ }
+
out.close();
}
try {
out << "numSampled\t";
- for (int i = 0; i < globaldata->Groups.size(); i++) { out << globaldata->Groups[i] << '\t'; }
+ for (int i = 0; i < m->Groups.size(); i++) { out << m->Groups[i] << '\t'; }
out << endl;
out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
out << numSampled << '\t';
- for (int j = 0; j < globaldata->Groups.size(); j++) {
- if (numSampled < div[globaldata->Groups[j]].size()) {
+ for (int j = 0; j < m->Groups.size(); j++) {
+ if (numSampled < div[m->Groups[j]].size()) {
float score;
- if (scale) { score = (div[globaldata->Groups[j]][numSampled] / (float)numIters) / (float)numSampled; }
- else { score = div[globaldata->Groups[j]][numSampled] / (float)numIters; }
+ if (scale) { score = (div[m->Groups[j]][numSampled] / (float)numIters) / (float)numSampled; }
+ else { score = div[m->Groups[j]][numSampled] / (float)numIters; }
out << setprecision(4) << score << '\t';
}else { out << "NA" << '\t'; }
}
}
//**********************************************************************************************************************
-float PhyloDiversityCommand::calcBranchLength(Tree* t, int leaf){
+//need a vector of floats one branch length for every group the node represents.
+vector<float> PhyloDiversityCommand::calcBranchLength(Tree* t, int leaf, map< string, set<int> >& counted){
try {
//calc the branch length
//while you aren't at root
- float sum = 0.0;
+ vector<float> sums;
int index = leaf;
-
+
+ vector<string> groups = t->tree[leaf].getGroup();
+ sums.resize(groups.size(), 0.0);
+
+ map<string, map<int, double> > tempTotals; //maps node to total Branch Length
+ map< string, set<int> > tempCounted;
+ set<int>::iterator it;
+
+ //you are a leaf
+ if(t->tree[index].getBranchLength() != -1){
+ for (int k = 0; k < groups.size(); k++) {
+ sums[k] += abs(t->tree[index].getBranchLength());
+ counted[groups[k]].insert(index);
+ }
+ }
+
+ for (int k = 0; k < groups.size(); k++) {
+ tempTotals[groups[k]][index] = 0.0;
+ }
+
+ index = t->tree[index].getParent();
+
+ //while you aren't at root
while(t->tree[index].getParent() != -1){
+
+ if (m->control_pressed) { return sums; }
+
+ int pcountSize = 0;
+ for (int k = 0; k < groups.size(); k++) {
+ map<string, int>::iterator itGroup = t->tree[index].pcount.find(groups[k]);
+ if (itGroup != t->tree[index].pcount.end()) { pcountSize++; }
- //if you have a BL
- if(t->tree[index].getBranchLength() != -1){
- sum += abs(t->tree[index].getBranchLength());
+ //do both your chidren have have descendants from the users groups?
+ int lc = t->tree[index].getLChild();
+ int rc = t->tree[index].getRChild();
+
+ int LpcountSize = 0;
+ itGroup = t->tree[lc].pcount.find(groups[k]);
+ if (itGroup != t->tree[lc].pcount.end()) { LpcountSize++; }
+
+ int RpcountSize = 0;
+ itGroup = t->tree[rc].pcount.find(groups[k]);
+ if (itGroup != t->tree[rc].pcount.end()) { RpcountSize++; }
+
+ //if yes, add your childrens tempTotals
+ if ((LpcountSize != 0) && (RpcountSize != 0)) {
+ sums[k] += tempTotals[groups[k]][lc] + tempTotals[groups[k]][rc];
+
+ for (it = tempCounted[groups[k]].begin(); it != tempCounted[groups[k]].end(); it++) { counted[groups[k]].insert(*it); }
+
+ //cout << "added to total " << tempTotals[lc] << '\t' << tempTotals[rc] << endl;
+ if (t->tree[index].getBranchLength() != -1) {
+ if (counted[groups[k]].count(index) == 0) {
+ tempTotals[groups[k]][index] = abs(t->tree[index].getBranchLength());
+ tempCounted[groups[k]].insert(index);
+ }else{
+ tempTotals[groups[k]][index] = 0.0;
+ }
+ }else {
+ tempTotals[groups[k]][index] = 0.0;
+ }
+ }else { //if no, your tempTotal is your childrens temp totals + your branch length
+ tempTotals[groups[k]][index] = tempTotals[groups[k]][lc] + tempTotals[groups[k]][rc];
+
+ if (counted[groups[k]].count(index) == 0) {
+ tempTotals[groups[k]][index] += abs(t->tree[index].getBranchLength());
+ tempCounted[groups[k]].insert(index);
+ }
+
+ }
+ //cout << "temptotal = "<< tempTotals[i] << endl;
}
- index = t->tree[index].getParent();
+
+ index = t->tree[index].getParent();
}
+
+ return sums;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloDiversityCommand", "calcBranchLength");
+ exit(1);
+ }
+}
+/*****************************************************************/
+int PhyloDiversityCommand::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);
- //get last breanch length added
- if(t->tree[index].getBranchLength() != -1){
- sum += abs(t->tree[index].getBranchLength());
+ 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 sum;
+ return 0;
}
catch(exception& e) {
- m->errorOut(e, "PhyloDiversityCommand", "calcBranchLength");
+ m->errorOut(e, "PhyloDiversityCommand", "readNamesFile");
exit(1);
}
}
-//**********************************************************************************************************************
\ No newline at end of file
+
+//**********************************************************************************************************************
+
+
+
projects / mothur.git / blobdiff