5 * Created by Sarah Westcott on 5/7/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
10 #include "distancecommand.h"
11 #include "ignoregaps.h"
12 #include "eachgapdist.h"
13 #include "eachgapignore.h"
14 #include "onegapdist.h"
15 #include "onegapignore.h"
17 //**********************************************************************************************************************
19 DistanceCommand::DistanceCommand(string option){
21 globaldata = GlobalData::getInstance();
25 //allow user to run help
26 if(option == "help") { help(); abort = true; }
29 //valid paramters for this command
30 string Array[] = {"fasta", "phylip", "calc", "countends", "cutoff", "processors"};
31 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
33 parser = new OptionParser();
34 parser->parse(option, parameters); delete parser;
36 ValidParameters* validParameter = new ValidParameters();
38 //check to make sure all parameters are valid for command
39 for (it2 = parameters.begin(); it2 != parameters.end(); it2++) {
40 if (validParameter->isValidParameter(it2->first, myArray, it2->second) != true) { abort = true; }
43 //check for required parameters
44 fastafile = validParameter->validFile(parameters, "fasta", true);
45 if (fastafile == "not found") { cout << "fasta is a required parameter for the dist.seqs command." << endl; abort = true; }
46 else if (fastafile == "not open") { abort = true; }
48 globaldata->setFastaFile(fastafile);
49 openInputFile(fastafile, in);
52 //check for optional parameter and set defaults
53 // ...at some point should added some additional type checking...
54 calc = validParameter->validFile(parameters, "calc", false);
55 if (calc == "not found") { calc = "onegap"; }
57 if (calc == "default") { calc = "onegap"; }
59 splitAtDash(calc, Estimators);
62 temp = validParameter->validFile(parameters, "countends", false); if (temp == "not found") { temp = "T"; }
63 convert(temp, countends);
65 temp = validParameter->validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "1.0"; }
66 convert(temp, cutoff);
68 temp = validParameter->validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
69 convert(temp, processors);
71 phylip = validParameter->validFile(parameters, "phylip", false); if (phylip == "not found") { phylip = "F"; }
73 delete validParameter;
75 validCalculator = new ValidCalculators();
78 if (isTrue(countends) == true) {
79 for (i=0; i<Estimators.size(); i++) {
80 if (validCalculator->isValidCalculator("distance", Estimators[i]) == true) {
81 if (Estimators[i] == "nogaps") {
82 distCalculator = new ignoreGaps();
83 }else if (Estimators[i] == "eachgap") {
84 distCalculator = new eachGapDist();
85 }else if (Estimators[i] == "onegap") {
86 distCalculator = new oneGapDist(); }
90 for (i=0; i<Estimators.size(); i++) {
91 if (validCalculator->isValidCalculator("distance", Estimators[i]) == true) {
92 if (Estimators[i] == "nogaps") {
93 distCalculator = new ignoreGaps();
94 }else if (Estimators[i] == "eachgap") {
95 distCalculator = new eachGapIgnoreTermGapDist();
96 }else if (Estimators[i] == "onegap") {
97 distCalculator = new oneGapIgnoreTermGapDist();
103 delete validCalculator;
107 catch(exception& e) {
108 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function DistanceCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
112 cout << "An unknown error has occurred in the DistanceCommand class function DistanceCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
116 //**********************************************************************************************************************
118 void DistanceCommand::help(){
120 cout << "The dist.seqs command reads a file containing sequences and creates a distance file." << "\n";
121 cout << "The dist.seqs command parameters are fasta, calc, countends, cutoff and processors. " << "\n";
122 cout << "The fasta parameter is required." << "\n";
123 cout << "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";
124 cout << "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";
125 cout << "The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0." << "\n";
126 cout << "The processors parameter allows you to specify number of processors to use. The default is 1." << "\n";
127 cout << "The dist.seqs command should be in the following format: " << "\n";
128 cout << "dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) " << "\n";
129 cout << "Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3)." << "\n";
130 cout << "Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc)." << "\n" << "\n";
132 catch(exception& e) {
133 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
137 cout << "An unknown error has occurred in the DistanceCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
143 //**********************************************************************************************************************
145 int DistanceCommand::execute(){
148 if (abort == true) { return 0; }
150 //reads fasta file and fills sequenceDB
151 seqDB = new SequenceDB(in);
153 int numSeqs = seqDB->getNumSeqs();
158 //doses the user want the phylip formatted file as well
159 if (isTrue(phylip) == true) {
160 outputFile = getRootName(fastafile) + "phylip.dist";
161 remove(outputFile.c_str());
163 //output numSeqs to phylip formatted dist file
164 openOutputFile(outputFile, outFile);
165 outFile << numSeqs << endl;
167 }else { //user wants column format
168 outputFile = getRootName(fastafile) + "dist";
169 remove(outputFile.c_str());
172 //# if defined (_WIN32)
173 //figure out how to implement the fork and wait commands in windows
174 // driver(distCalculator, seqDB, 0, numSeqs, distFile, phylipFile, cutoff);
178 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
179 //if you don't need to fork anything
181 driver(distCalculator, seqDB, 0, numSeqs, outputFile + ".temp", cutoff);
182 appendFiles((outputFile + ".temp"), outputFile);
183 remove((outputFile + ".temp").c_str());
184 }else{ //you have multiple processors
186 for (int i = 0; i < processors; i++) {
187 lines.push_back(new linePair());
188 lines[i]->start = int (sqrt(float(i)/float(processors)) * numSeqs);
189 lines[i]->end = int (sqrt(float(i+1)/float(processors)) * numSeqs);
192 cout << lines[0]->start << '\t' << lines[0]->end << endl;
193 cout << lines[1]->start << '\t' << lines[1]->end << endl;
195 createProcesses(outputFile);
197 //append and remove temp files
198 for (it = processIDS.begin(); it != processIDS.end(); it++) {
199 appendFiles((outputFile + toString(it->second) + ".temp"), outputFile);
200 remove((outputFile + toString(it->second) + ".temp").c_str());
204 driver(distCalculator, seqDB, 0, numSeqs, outputFile + ".temp", cutoff);
205 appendFiles((outputFile + ".temp"), outputFile);
206 remove((outputFile + ".temp").c_str());
209 delete distCalculator;
214 catch(exception& e) {
215 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
219 cout << "An unknown error has occurred in the DistanceCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
223 /**************************************************************************************************/
224 void DistanceCommand::createProcesses(string filename) {
229 //loop through and create all the processes you want
230 while (process != processors) {
234 processIDS[lines[process]->end] = pid; //create map from line number to pid so you can append files in correct order later
237 driver(distCalculator, seqDB, lines[process]->start, lines[process]->end, filename + toString(getpid()) + ".temp", cutoff);
239 }else { cout << "unable to spawn the necessary processes." << endl; exit(0); }
242 //force parent to wait until all the processes are done
243 for (it = processIDS.begin(); it != processIDS.end(); it++) {
244 int temp = it->second;
249 catch(exception& e) {
250 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function createProcesses. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
254 cout << "An unknown error has occurred in the DistanceCommand class function createProcesses. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
259 /**************************************************************************************************/
260 /////// need to fix to work with calcs and sequencedb
261 int DistanceCommand::driver(Dist* distCalculator, SequenceDB* align, int startLine, int endLine, string dFileName, float cutoff){
264 int startTime = time(NULL);
267 ofstream outFile(dFileName.c_str(), ios::trunc);
268 outFile.setf(ios::fixed, ios::showpoint);
269 outFile << setprecision(4);
271 for(int i=startLine;i<endLine;i++){
273 for(int j=0;j<i;j++){
274 distCalculator->calcDist(*(align->get(i)), *(align->get(j)));
275 double dist = distCalculator->getDist();
278 if (isTrue(phylip) != true) { outFile << align->get(i)->getName() << ' ' << align->get(j)->getName() << ' ' << dist << endl; }
280 if (isTrue(phylip) == true) { outFile << dist << '\t'; }
284 if (isTrue(phylip) == true) { outFile << endl; }
287 cout << i << '\t' << time(NULL) - startTime << endl;
291 cout << endLine-1 << '\t' << time(NULL) - startTime << endl;
298 catch(exception& e) {
299 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function driver. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
303 cout << "An unknown error has occurred in the DistanceCommand class function driver. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
309 /**************************************************************************************************/
310 void DistanceCommand::appendFiles(string temp, string filename) {
315 //open output file in append mode
316 openOutputFileAppend(filename, output);
318 //open temp file for reading
319 openInputFile(temp, input);
322 //read input file and write to output file
323 while(input.eof() != true) {
324 getline(input, line); //getline removes the newline char
326 output << line << endl; // Appending back newline char
333 catch(exception& e) {
334 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
338 cout << "An unknown error has occurred in the DistanceCommand class function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
342 /**************************************************************************************************/