2 * chimerabellerophoncommand.cpp
5 * Created by westcott on 4/1/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "chimerabellerophoncommand.h"
11 #include "bellerophon.h"
13 //***************************************************************************************************************
15 ChimeraBellerophonCommand::ChimeraBellerophonCommand(string option) {
19 //allow user to run help
20 if(option == "help") { help(); abort = true; }
23 //valid paramters for this command
24 string Array[] = {"fasta","filter","correction","processors","window","increment","outputdir","inputdir"};
25 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
27 OptionParser parser(option);
28 map<string,string> parameters = parser.getParameters();
30 ValidParameters validParameter;
31 map<string,string>::iterator it;
33 //check to make sure all parameters are valid for command
34 for (it = parameters.begin(); it != parameters.end(); it++) {
35 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
38 //if the user changes the input directory command factory will send this info to us in the output parameter
39 string inputDir = validParameter.validFile(parameters, "inputdir", false);
40 if (inputDir == "not found"){ inputDir = ""; }
42 fastafile = validParameter.validFile(parameters, "fasta", false);
43 if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.bellerophon command."); m->mothurOutEndLine(); abort = true; }
45 splitAtDash(fastafile, fastaFileNames);
47 //go through files and make sure they are good, if not, then disregard them
48 for (int i = 0; i < fastaFileNames.size(); i++) {
50 string path = hasPath(fastaFileNames[i]);
51 //if the user has not given a path then, add inputdir. else leave path alone.
52 if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
60 MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
61 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
66 ableToOpen = openInputFile(fastaFileNames[i], in, "noerror");
68 //if you can't open it, try default location
69 if (ableToOpen == 1) {
70 if (m->getDefaultPath() != "") { //default path is set
71 string tryPath = m->getDefaultPath() + getSimpleName(fastaFileNames[i]);
72 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
73 ableToOpen = openInputFile(tryPath, in, "noerror");
74 fastaFileNames[i] = tryPath;
80 for (int j = 1; j < processors; j++) {
81 MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
85 MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
90 if (ableToOpen == 1) {
91 m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
92 //erase from file list
93 fastaFileNames.erase(fastaFileNames.begin()+i);
99 //make sure there is at least one valid file left
100 if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; }
103 //if the user changes the output directory command factory will send this info to us in the output parameter
104 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
106 outputDir += hasPath(fastafile); //if user entered a file with a path then preserve it
110 temp = validParameter.validFile(parameters, "filter", false); if (temp == "not found") { temp = "F"; }
111 filter = isTrue(temp);
113 temp = validParameter.validFile(parameters, "correction", false); if (temp == "not found") { temp = "T"; }
114 correction = isTrue(temp);
116 temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
117 convert(temp, processors);
119 temp = validParameter.validFile(parameters, "window", false); if (temp == "not found") { temp = "0"; }
120 convert(temp, window);
122 temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "25"; }
123 convert(temp, increment);
126 catch(exception& e) {
127 m->errorOut(e, "ChimeraBellerophonCommand", "ChimeraBellerophonCommand");
131 //**********************************************************************************************************************
133 void ChimeraBellerophonCommand::help(){
135 m->mothurOut("The chimera.bellerophon command reads a fastafile and creates list of potentially chimeric sequences.\n");
136 m->mothurOut("The chimera.bellerophon command parameters are fasta, filter, correction, processors, window, increment. The fasta parameter is required.\n");
137 m->mothurOut("The fasta parameter is required. You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n");
138 m->mothurOut("The filter parameter allows you to specify if you would like to apply a vertical and 50% soft filter, default=false. \n");
139 m->mothurOut("The correction parameter allows you to put more emphasis on the distance between highly similar sequences and less emphasis on the differences between remote homologs, default=true.\n");
140 m->mothurOut("The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n");
142 m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n");
144 m->mothurOut("The window parameter allows you to specify the window size for searching for chimeras, default is 1/4 sequence length. \n");
145 m->mothurOut("The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default is 25.\n");
146 m->mothurOut("chimera.bellerophon(fasta=yourFastaFile, filter=yourFilter, correction=yourCorrection, processors=yourProcessors) \n");
147 m->mothurOut("Example: chimera.bellerophon(fasta=AD.align, filter=True, correction=true, window=200) \n");
148 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
150 catch(exception& e) {
151 m->errorOut(e, "ChimeraBellerophonCommand", "help");
156 //***************************************************************************************************************
158 ChimeraBellerophonCommand::~ChimeraBellerophonCommand(){ /* do nothing */ }
160 //***************************************************************************************************************
162 int ChimeraBellerophonCommand::execute(){
165 if (abort == true) { return 0; }
167 for (int i = 0; i < fastaFileNames.size(); i++) {
169 m->mothurOut("Checking sequences from " + fastaFileNames[i] + " ..." ); m->mothurOutEndLine();
171 int start = time(NULL);
173 chimera = new Bellerophon(fastaFileNames[i], filter, correction, window, increment, processors, outputDir);
175 string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "bellerophon.chimeras";
176 string accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "bellerophon.accnos";
178 chimera->getChimeras();
180 if (m->control_pressed) { delete chimera; for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
184 MPI_File outMPIAccnos;
186 int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
188 char outFilename[1024];
189 strcpy(outFilename, accnosFileName.c_str());
192 strcpy(FileName, outputFileName.c_str());
194 MPI_File_open(MPI_COMM_WORLD, FileName, outMode, MPI_INFO_NULL, &outMPI); //comm, filename, mode, info, filepointer
195 MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPIAccnos);
197 numSeqs = chimera->print(outMPI, outMPIAccnos);
199 MPI_File_close(&outMPI);
200 MPI_File_close(&outMPIAccnos);
205 openOutputFile(outputFileName, out);
208 openOutputFile(accnosFileName, out2);
210 numSeqs = chimera->print(out, out2);
216 if (m->control_pressed) { remove(accnosFileName.c_str()); remove(outputFileName.c_str()); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete chimera; return 0; }
218 m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
220 outputNames.push_back(outputFileName);
221 outputNames.push_back(accnosFileName);
226 m->mothurOutEndLine();
227 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
228 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
229 m->mothurOutEndLine();
234 catch(exception& e) {
235 m->errorOut(e, "ChimeraBellerophonCommand", "execute");
239 //**********************************************************************************************************************