vector<int> LefseCommand::runKruskalWallis(vector<SharedRAbundVector*>& lookup, DesignMap& designMap) {
try {
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile));
variables["[distance]"] = lookup[0]->getLabel();
string outputFileName = getOutputFileName("kruskall-wallis",variables);
ofstream out;
m->openOutputFile(outputFileName, out);
outputNames.push_back(outputFileName); outputTypes["kruskall-wallis"].push_back(outputFileName);
out << "OTULabel\tKW\tPvalue\n";
vector<int> significantOtuLabels;
int numBins = lookup[0]->getNumBins();
//sanity check to make sure each treatment has a group in the shared file
set<string> treatments;
for (int j = 0; j < lookup.size(); j++) {
string group = lookup[j]->getGroup();
string treatment = designMap.get(group, mclass); //get value for this group in this category
treatments.insert(treatment);
}
if (treatments.size() < 2) { m->mothurOut("[ERROR]: need at least 2 things to classes to compare, quitting.\n"); m->control_pressed = true; }
LinearAlgebra linear;
for (int i = 0; i < numBins; i++) {
if (m->control_pressed) { break; }
vector<spearmanRank> values;
for (int j = 0; j < lookup.size(); j++) {
string group = lookup[j]->getGroup();
string treatment = designMap.get(group, mclass); //get value for this group in this category
spearmanRank temp(treatment, lookup[j]->getAbundance(i));
values.push_back(temp);
}
double pValue = 0.0;
double H = linear.calcKruskalWallis(values, pValue);
//output H and signifigance
out << m->currentBinLabels[i] << '\t' << H << '\t' << pValue << endl;
if (pValue < anovaAlpha) { significantOtuLabels.push_back(i); }
}
out.close();
return significantOtuLabels;
}
catch(exception& e) {
m->errorOut(e, "LefseCommand", "runKruskalWallis");
exit(1);
}
}
qFinderDMM::qFinderDMM(vector<vector<int> > cm, int p): countMatrix(cm), numPartitions(p){
try {
m = MothurOut::getInstance();
numSamples = (int)countMatrix.size();
numOTUs = (int)countMatrix[0].size();
kMeans();
// cout << "done kMeans" << endl;
optimizeLambda();
// cout << "done optimizeLambda" << endl;
double change = 1.0000;
currNLL = 0.0000;
int iter = 0;
while(change > 1.0e-6 && iter < 100){
// cout << "Calc_Z: ";
calculatePiK();
optimizeLambda();
// printf("Iter:%d\t",iter);
for(int i=0;i<numPartitions;i++){
weights[i] = 0.0000;
for(int j=0;j<numSamples;j++){
weights[i] += zMatrix[i][j];
}
// printf("w_%d=%.3f\t",i,weights[i]);
}
double nLL = getNegativeLogLikelihood();
change = abs(nLL - currNLL);
currNLL = nLL;
// printf("NLL=%.5f\tDelta=%.4e\n",currNLL, change);
iter++;
}
error.resize(numPartitions);
logDeterminant = 0.0000;
LinearAlgebra l;
for(currentPartition=0;currentPartition<numPartitions;currentPartition++){
error[currentPartition].assign(numOTUs, 0.0000);
if(currentPartition > 0){
logDeterminant += (2.0 * log(numSamples) - log(weights[currentPartition]));
}
vector<vector<double> > hessian = getHessian();
vector<vector<double> > invHessian = l.getInverse(hessian);
for(int i=0;i<numOTUs;i++){
logDeterminant += log(abs(hessian[i][i]));
error[currentPartition][i] = invHessian[i][i];
}
}
int numParameters = numPartitions * numOTUs + numPartitions - 1;
laplace = currNLL + 0.5 * logDeterminant - 0.5 * numParameters * log(2.0 * 3.14159);
bic = currNLL + 0.5 * log(numSamples) * numParameters;
aic = currNLL + numParameters;
}
catch(exception& e) {
m->errorOut(e, "qFinderDMM", "qFinderDMM");
exit(1);
}
}
qFinderDMM::qFinderDMM(vector<vector<int> > cm, int p) : CommunityTypeFinder() {
try {
//cout << "here" << endl;
numPartitions = p;
countMatrix = cm;
numSamples = (int)countMatrix.size();
numOTUs = (int)countMatrix[0].size();
// if (m->debug) { m->mothurOut("before kmeans\n"); }
findkMeans();
//if (m->debug) { m->mothurOut("done kMeans\n"); }
optimizeLambda();
//if (m->debug) { m->mothurOut("done optimizeLambda\n"); }
double change = 1.0000;
currNLL = 0.0000;
int iter = 0;
while(change > 1.0e-6 && iter < 100){
// if (m->debug) { m->mothurOut("Calc_Z: \n"); }
calculatePiK();
optimizeLambda();
// if (m->debug) { m->mothurOut("Iter: " + toString(iter) + "\n"); }
for(int i=0;i<numPartitions;i++){
weights[i] = 0.0000;
for(int j=0;j<numSamples;j++){
weights[i] += zMatrix[i][j];
}
// printf("w_%d=%.3f\t",i,weights[i]);
}
double nLL = getNegativeLogLikelihood();
change = abs(nLL - currNLL);
currNLL = nLL;
// printf("NLL=%.5f\tDelta=%.4e\n",currNLL, change);
iter++;
}
//if (m->debug) { m->mothurOut("done while loop\n"); }
error.resize(numPartitions);
logDeterminant = 0.0000;
LinearAlgebra l;
for(currentPartition=0;currentPartition<numPartitions;currentPartition++){
error[currentPartition].assign(numOTUs, 0.0000);
if (m->debug) { m->mothurOut("current partition = " + toString(currentPartition) + "\n"); }
if(currentPartition > 0){
logDeterminant += (2.0 * log(numSamples) - log(weights[currentPartition]));
}
//if (m->debug) { m->mothurOut("before hession\n"); }
vector<vector<double> > hessian = getHessian();
//if (m->debug) { m->mothurOut("after hession\n"); }
vector<vector<double> > invHessian = l.getInverse(hessian);
//if (m->debug) { m->mothurOut("after inverse\n"); }
for(int i=0;i<numOTUs;i++){
logDeterminant += log(abs(hessian[i][i]));
error[currentPartition][i] = invHessian[i][i];
}
}
int numParameters = numPartitions * numOTUs + numPartitions - 1;
laplace = currNLL + 0.5 * logDeterminant - 0.5 * numParameters * log(2.0 * 3.14159);
bic = currNLL + 0.5 * log(numSamples) * numParameters;
aic = currNLL + numParameters;
}
catch(exception& e) {
m->errorOut(e, "qFinderDMM", "qFinderDMM");
exit(1);
}
}
vector<int> LefseCommand::runWilcoxon(vector<SharedRAbundVector*>& lookup, DesignMap& designMap, vector<int> bins) {
try {
vector<int> significantOtuLabels;
//if it exists and meets the following requirements run Wilcoxon
/*
1. Subclass members all belong to same main class
2. Number of groups in each subclass is the same
3. anything else??
*/
vector<string> subclasses;
map<string, string> subclass2Class;
map<string, int> subclassCounts;
map<string, vector<int> > subClass2GroupIndex; //maps subclass name to vector of indexes in lookup from that subclass. old -> 1,2,3 means groups in location 1,2,3 of lookup are from old. Saves time below.
bool error = false;
for (int j = 0; j < lookup.size(); j++) {
string group = lookup[j]->getGroup();
string treatment = designMap.get(group, mclass); //get value for this group in this category
string thisSub = designMap.get(group, subclass);
map<string, string>::iterator it = subclass2Class.find(thisSub);
if (it == subclass2Class.end()) {
subclass2Class[thisSub] = treatment;
subclassCounts[thisSub] = 1;
vector<int> temp; temp.push_back(j);
subClass2GroupIndex[thisSub] = temp;
}
else {
subclassCounts[thisSub]++;
subClass2GroupIndex[thisSub].push_back(j);
if (it->second != treatment) {
error = true;
m->mothurOut("[ERROR]: subclass " + thisSub + " has members in " + it->second + " and " + treatment + ". Subclass members must be from the same class. Ignoring wilcoxon.\n");
}
}
}
if (error) { return significantOtuLabels; }
else { //check counts to make sure subclasses are the same size
set<int> counts;
for (map<string, int>::iterator it = subclassCounts.begin(); it != subclassCounts.end(); it++) { counts.insert(it->second); }
if (counts.size() > 1) { m->mothurOut("[ERROR]: subclasses must be the same size. Ignoring wilcoxon.\n");
return significantOtuLabels; }
}
int numBins = lookup[0]->getNumBins();
vector<compGroup> comp;
//find comparisons and fill comp
map<string, int>::iterator itB;
for(map<string, int>::iterator it=subclassCounts.begin();it!=subclassCounts.end();it++){
itB = it;itB++;
for(itB;itB!=subclassCounts.end();itB++){
compGroup temp(it->first,itB->first);
comp.push_back(temp);
}
}
int numComp = comp.size();
if (numComp < 2) { m->mothurOut("[ERROR]: Need at least 2 subclasses, Ignoring Wilcoxon.\n");
return significantOtuLabels; }
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile));
variables["[distance]"] = lookup[0]->getLabel();
string outputFileName = getOutputFileName("wilcoxon",variables);
ofstream out;
m->openOutputFile(outputFileName, out);
outputNames.push_back(outputFileName); outputTypes["wilcoxon"].push_back(outputFileName);
out << "OTULabel\tComparision\tWilcoxon\tPvalue\n";
LinearAlgebra linear;
for (int i = 0; i < numBins; i++) {
if (m->control_pressed) { break; }
if (m->inUsersGroups(i, bins)) { //flagged in Kruskal Wallis
bool sig = false;
//for each subclass comparision
for (int j = 0; j < numComp; j++) {
//fill x and y with this comparisons data
vector<double> x; vector<double> y;
//fill x and y
vector<int> xIndexes = subClass2GroupIndex[comp[j].group1]; //indexes in lookup for this subclass
for (int k = 0; k < xIndexes.size(); k++) { x.push_back(lookup[xIndexes[k]]->getAbundance(i)); }
vector<int> yIndexes = subClass2GroupIndex[comp[j].group2]; //indexes in lookup for this subclass
for (int k = 0; k < yIndexes.size(); k++) { y.push_back(lookup[yIndexes[k]]->getAbundance(i)); }
double pValue = 0.0;
double H = linear.calcWilcoxon(x, y, pValue);
//output H and signifigance
out << m->currentBinLabels[i] << '\t' << comp[j].getCombo() << '\t' << H << '\t' << pValue << endl;
//set sig - not sure how yet
}
if (sig) { significantOtuLabels.push_back(i); }
}
}
out.close();
return significantOtuLabels;
}
catch(exception& e) {
m->errorOut(e, "LefseCommand", "runWilcoxon");
exit(1);
}
}
