//**********************************************************************************************************************
-DistanceCommand::DistanceCommand(string option){
+DistanceCommand::DistanceCommand(string option) {
try {
abort = false;
Estimators.clear();
-
+
//allow user to run help
if(option == "help") { help(); abort = true; }
else {
//valid paramters for this command
- string Array[] = {"fasta", "phylip", "calc", "countends", "cutoff", "processors"};
+ string Array[] = {"fasta", "output", "calc", "countends", "cutoff", "processors", "outputdir","inputdir"};
vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
ValidParameters validParameter;
+ map<string, string>::iterator it2;
//check to make sure all parameters are valid for command
- for (map<string, string>::iterator it2 = parameters.begin(); it2 != parameters.end(); it2++) {
+ for (it2 = parameters.begin(); it2 != parameters.end(); it2++) {
if (validParameter.isValidParameter(it2->first, myArray, it2->second) != true) { abort = true; }
}
+ //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;
+ it2 = parameters.find("fasta");
+ //user has given a template file
+ if(it2 != parameters.end()){
+ path = hasPath(it2->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["fasta"] = inputDir + it2->second; }
+ }
+ }
+
//check for required parameters
fastafile = validParameter.validFile(parameters, "fasta", true);
- if (fastafile == "not found") { mothurOut("fasta is a required parameter for the dist.seqs command."); mothurOutEndLine(); abort = true; }
+ if (fastafile == "not found") { m->mothurOut("fasta is a required parameter for the dist.seqs command."); m->mothurOutEndLine(); abort = true; }
else if (fastafile == "not open") { abort = true; }
else{
ifstream inFASTA;
alignDB = SequenceDB(inFASTA);
inFASTA.close();
}
+
+ //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 += hasPath(fastafile); //if user entered a file with a path then preserve it
+ }
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
temp = validParameter.validFile(parameters, "processors", false); if(temp == "not found"){ temp = "1"; }
convert(temp, processors);
- phylip = validParameter.validFile(parameters, "phylip", false); if(phylip == "not found"){ phylip = "F"; }
-
+ output = validParameter.validFile(parameters, "output", false); if(output == "not found"){ output = "column"; }
+
+ if ((output != "column") && (output != "lt") && (output != "square")) { m->mothurOut(output + " is not a valid output form. Options are column, lt and square. I will use column."); m->mothurOutEndLine(); output = "column"; }
ValidCalculators validCalculator;
}
catch(exception& e) {
- errorOut(e, "DistanceCommand", "DistanceCommand");
+ m->errorOut(e, "DistanceCommand", "DistanceCommand");
exit(1);
}
}
void DistanceCommand::help(){
try {
- mothurOut("The dist.seqs command reads a file containing sequences and creates a distance file.\n");
- mothurOut("The dist.seqs command parameters are fasta, calc, countends, cutoff and processors. \n");
- mothurOut("The fasta parameter is required.\n");
- mothurOut("The calc parameter allows you to specify the method of calculating the distances. Your options are: nogaps, onegap or eachgap. The default is onegap.\n");
- mothurOut("The countends parameter allows you to specify whether to include terminal gaps in distance. Your options are: T or F. The default is T.\n");
- mothurOut("The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0.\n");
- mothurOut("The processors parameter allows you to specify number of processors to use. The default is 1.\n");
- mothurOut("The dist.seqs command should be in the following format: \n");
- mothurOut("dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) \n");
- mothurOut("Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3).\n");
- mothurOut("Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc).\n\n");
+ m->mothurOut("The dist.seqs command reads a file containing sequences and creates a distance file.\n");
+ m->mothurOut("The dist.seqs command parameters are fasta, calc, countends, output, cutoff and processors. \n");
+ m->mothurOut("The fasta parameter is required.\n");
+ m->mothurOut("The calc parameter allows you to specify the method of calculating the distances. Your options are: nogaps, onegap or eachgap. The default is onegap.\n");
+ m->mothurOut("The countends parameter allows you to specify whether to include terminal gaps in distance. Your options are: T or F. The default is T.\n");
+ m->mothurOut("The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0.\n");
+ m->mothurOut("The output parameter allows you to specify format of your distance matrix. Options are column, lt, and square. The default is column.\n");
+ m->mothurOut("The processors parameter allows you to specify number of processors to use. The default is 1.\n");
+ m->mothurOut("The dist.seqs command should be in the following format: \n");
+ m->mothurOut("dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) \n");
+ m->mothurOut("Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3).\n");
+ m->mothurOut("Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc).\n\n");
}
catch(exception& e) {
- errorOut(e, "DistanceCommand", "help");
+ m->errorOut(e, "DistanceCommand", "help");
exit(1);
}
}
string outputFile;
- //doses the user want the phylip formatted file as well
- if (isTrue(phylip) == true) {
- outputFile = getRootName(fastafile) + "phylip.dist";
+ if (output == "lt") { //does the user want lower triangle phylip formatted file
+ outputFile = outputDir + getRootName(getSimpleName(fastafile)) + "phylip.dist";
remove(outputFile.c_str());
//output numSeqs to phylip formatted dist file
- }else { //user wants column format
- outputFile = getRootName(fastafile) + "dist";
+ }else if (output == "column") { //user wants column format
+ outputFile = outputDir + getRootName(getSimpleName(fastafile)) + "dist";
+ remove(outputFile.c_str());
+ }else { //assume square
+ outputFile = outputDir + getRootName(getSimpleName(fastafile)) + "square.dist";
remove(outputFile.c_str());
}
- //# if defined (_WIN32)
- //figure out how to implement the fork and wait commands in windows
- // driver(distCalculator, seqDB, 0, numSeqs, distFile, phylipFile, cutoff);
- //# endif
-
-
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
//if you don't need to fork anything
if(processors == 1){
ifstream inFASTA;
driver(0, numSeqs, outputFile, cutoff);
#endif
+ if (m->control_pressed) { delete distCalculator; remove(outputFile.c_str()); return 0; }
+
+ if (output == "square") { convertMatrix(outputFile); }
+
+ if (m->control_pressed) { delete distCalculator; remove(outputFile.c_str()); return 0; }
delete distCalculator;
+ m->mothurOutEndLine();
+ m->mothurOut("Output File Name: "); m->mothurOutEndLine();
+ m->mothurOut(outputFile); m->mothurOutEndLine();
+ m->mothurOutEndLine();
+
return 0;
}
catch(exception& e) {
- errorOut(e, "DistanceCommand", "execute");
+ m->errorOut(e, "DistanceCommand", "execute");
exit(1);
}
}
}else if (pid == 0){
driver(lines[process]->start, lines[process]->end, filename + toString(getpid()) + ".temp", cutoff);
exit(0);
- }else { mothurOut("unable to spawn the necessary processes."); mothurOutEndLine(); exit(0); }
+ }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
}
//force parent to wait until all the processes are done
#endif
}
catch(exception& e) {
- errorOut(e, "DistanceCommand", "createProcesses");
+ m->errorOut(e, "DistanceCommand", "createProcesses");
exit(1);
}
}
outFile.setf(ios::fixed, ios::showpoint);
outFile << setprecision(4);
- if(isTrue(phylip) && startLine == 0){ outFile << alignDB.getNumSeqs() << endl; }
+ if((output == "lt") && startLine == 0){ outFile << alignDB.getNumSeqs() << endl; }
for(int i=startLine;i<endLine;i++){
- if(isTrue(phylip)) {
+ if(output == "lt") {
string name = alignDB.get(i).getName();
if (name.length() < 10) { //pad with spaces to make compatible
while (name.length() < 10) { name += " "; }
outFile << name << '\t';
}
for(int j=0;j<i;j++){
+
+ if (m->control_pressed) { outFile.close(); return 0; }
+
distCalculator->calcDist(alignDB.get(i), alignDB.get(j));
double dist = distCalculator->getDist();
if(dist <= cutoff){
- if (!isTrue(phylip)) { outFile << alignDB.get(i).getName() << ' ' << alignDB.get(j).getName() << ' ' << dist << endl; }
+ if (output == "column") { outFile << alignDB.get(i).getName() << ' ' << alignDB.get(j).getName() << ' ' << dist << endl; }
}
- if (isTrue(phylip)) { outFile << dist << '\t'; }
+ if (output == "lt") { outFile << dist << '\t'; }
+ if (output == "square") { //make a square column you can convert to square phylip
+ outFile << alignDB.get(i).getName() << '\t' << alignDB.get(j).getName() << '\t' << dist << endl;
+ outFile << alignDB.get(j).getName() << '\t' << alignDB.get(i).getName() << '\t' << dist << endl;
+ }
+
}
- if (isTrue(phylip) == true) { outFile << endl; }
+ if (output == "lt") { outFile << endl; }
if(i % 100 == 0){
- mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); mothurOutEndLine();
+ m->mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine();
}
}
- mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); mothurOutEndLine();
+ m->mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine();
outFile.close();
return 1;
}
catch(exception& e) {
- errorOut(e, "DistanceCommand", "driver");
+ m->errorOut(e, "DistanceCommand", "driver");
exit(1);
}
}
+/**************************************************************************************************/
+int DistanceCommand::convertMatrix(string outputFile) {
+ try{
+
+ //sort file by first column so the distances for each row are together
+ string outfile = getRootName(outputFile) + "sorted.dist.temp";
+
+ //use the unix sort
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ string command = "sort -n " + outputFile + " -o " + outfile;
+ system(command.c_str());
+ #else //sort using windows sort
+ string command = "sort " + outputFile + " /O " + outfile;
+ system(command.c_str());
+ #endif
+
+
+ //output to new file distance for each row and save positions in file where new row begins
+ ifstream in;
+ openInputFile(outfile, in);
+
+ ofstream out;
+ openOutputFile(outputFile, out);
+
+ out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
+ out << alignDB.getNumSeqs() << endl;
+
+ //get first currentRow
+ string first, currentRow, second;
+ float dist;
+ map<string, float> rowDists; //take advantage of the fact that maps are already sorted by key
+ map<string, float>::iterator it;
+
+ in >> first;
+ currentRow = first;
+
+ rowDists[first] = 0.00; //distance to yourself is 0.0
+
+ in.seekg(0);
+ //openInputFile(outfile, in);
+
+ while(!in.eof()) {
+ if (m->control_pressed) { in.close(); remove(outfile.c_str()); out.close(); return 0; }
+
+ in >> first >> second >> dist; gobble(in);
+
+ if (first != currentRow) {
+ //print out last row
+ out << currentRow << '\t'; //print name
+
+ //print dists
+ for (it = rowDists.begin(); it != rowDists.end(); it++) {
+ out << it->second << '\t';
+ }
+ out << endl;
+
+ //start new row
+ currentRow = first;
+ rowDists.clear();
+ rowDists[first] = 0.00;
+ rowDists[second] = dist;
+ }else{
+ rowDists[second] = dist;
+ }
+ }
+ //print out last row
+ out << currentRow << '\t'; //print name
+
+ //print dists
+ for (it = rowDists.begin(); it != rowDists.end(); it++) {
+ out << it->second << '\t';
+ }
+ out << endl;
+
+ in.close();
+ out.close();
+
+ remove(outfile.c_str());
+
+ return 1;
+
+ }
+ catch(exception& e) {
+ m->errorOut(e, "DistanceCommand", "convertMatrix");
+ exit(1);
+ }
+}
/**************************************************************************************************
void DistanceCommand::appendFiles(string temp, string filename) {
try{
output.close();
}
catch(exception& e) {
- errorOut(e, "DistanceCommand", "appendFiles");
+ m->errorOut(e, "DistanceCommand", "appendFiles");
exit(1);
}
}
/**************************************************************************************************/
+
+