]> git.donarmstrong.com Git - mothur.git/blobdiff - filterseqscommand.cpp
added name option to read.tree for use in unifrac and parimony commands
[mothur.git] / filterseqscommand.cpp
index ad491aa1252ccb28553b12daf0f353533428d531..8c61ab221d14b27e0f158b93189148817c72b057 100644 (file)
@@ -21,23 +21,52 @@ FilterSeqsCommand::FilterSeqsCommand(string option){
                
                else {
                        //valid paramters for this command
-                       string Array[] =  {"fasta", "trump", "soft", "hard", "vertical"};
+                       string Array[] =  {"fasta", "trump", "soft", "hard", "vertical", "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 it;
                        
                        //check to make sure all parameters are valid for command
-                       for (map<string,string>::iterator it = parameters.begin(); it != parameters.end(); it++) { 
+                       for (it = parameters.begin(); it != parameters.end(); it++) { 
                                if (validParameter.isValidParameter(it->first, myArray, it->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;
+                               it = parameters.find("fasta");
+                               //user has given a template file
+                               if(it != parameters.end()){ 
+                                       path = hasPath(it->second);
+                                       //if the user has not given a path then, add inputdir. else leave path alone.
+                                       if (path == "") {       parameters["fasta"] = inputDir + it->second;            }
+                               }
+                               
+                               it = parameters.find("hard");
+                               //user has given a template file
+                               if(it != parameters.end()){ 
+                                       path = hasPath(it->second);
+                                       //if the user has not given a path then, add inputdir. else leave path alone.
+                                       if (path == "") {       parameters["hard"] = inputDir + it->second;             }
+                               }
+                       }
+                       
                        //check for required parameters
                        fastafile = validParameter.validFile(parameters, "fasta", true);
                        if (fastafile == "not found") { mothurOut("fasta is a required parameter for the filter.seqs command."); mothurOutEndLine(); abort = true; }
                        else if (fastafile == "not open") { 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 = ""; 
+                               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...
@@ -56,6 +85,13 @@ FilterSeqsCommand::FilterSeqsCommand(string option){
                        
                        numSeqs = 0;
                        
+                       if (abort == false) {
+                       
+                               if (soft != 0)                  {  F.setSoft(soft);             }
+                               if (trump != '*')               {  F.setTrump(trump);   }
+                       
+                       }
+                                               
                }
                
        }
@@ -90,70 +126,6 @@ void FilterSeqsCommand::help(){
 
 /**************************************************************************************/
 
-void FilterSeqsCommand::doHard() {
-       
-       ifstream fileHandle;
-       openInputFile(hard, fileHandle);
-       
-       fileHandle >> filter;
-       
-       fileHandle.close();
-
-}
-
-/**************************************************************************************/
-
-void FilterSeqsCommand::doTrump(Sequence seq) {
-       
-       string curAligned = seq.getAligned();
-
-       for(int j = 0; j < alignmentLength; j++) {
-               if(curAligned[j] == trump){
-                       filter[j] = '0';
-               }
-       }
-
-}
-
-/**************************************************************************************/
-
-void FilterSeqsCommand::doVertical() {
-
-       for(int i=0;i<alignmentLength;i++){
-               if(gap[i] == numSeqs)   {       filter[i] = '0';        }
-       }
-       
-}
-
-/**************************************************************************************/
-
-void FilterSeqsCommand::doSoft() {
-       
-       int threshold = int (soft * numSeqs);
-       
-       for(int i=0;i<alignmentLength;i++){
-               if(a[i] < threshold && t[i] < threshold && g[i] < threshold && c[i] < threshold){       filter[i] = 0;  }
-       }
-}
-
-/**************************************************************************************/
-
-void FilterSeqsCommand::getFreqs(Sequence seq) {
-
-       string curAligned = seq.getAligned();;
-       
-       for(int j=0;j<alignmentLength;j++){
-               if(toupper(curAligned[j]) == 'A')                                                                               {       a[j]++;         }
-               else if(toupper(curAligned[j]) == 'T' || toupper(curAligned[j]) == 'U') {       t[j]++;         }
-               else if(toupper(curAligned[j]) == 'G')                                                                  {       g[j]++;         }
-               else if(toupper(curAligned[j]) == 'C')                                                                  {       c[j]++;         }
-               else if(curAligned[j] == '-' || curAligned[j] == '.')                                   {       gap[j]++;       }
-       }
-       
-}
-
-/**************************************************************************************/
-
 int FilterSeqsCommand::execute() {     
        try {
        
@@ -166,58 +138,64 @@ int FilterSeqsCommand::execute() {
                alignmentLength = testSeq.getAlignLength();
                inFASTA.seekg(0);
                
+               F.setLength(alignmentLength);
+               
                if(soft != 0 || isTrue(vertical)){
-                       a.assign(alignmentLength, 0);
-                       t.assign(alignmentLength, 0);
-                       g.assign(alignmentLength, 0);
-                       c.assign(alignmentLength, 0);
-                       gap.assign(alignmentLength, 0);
+                       F.initialize();
                }
                
-               if(hard.compare("") != 0)       {       doHard();               }
-               else                                            {       filter = string(alignmentLength, '1');  }
+               if(hard.compare("") != 0)       {       F.doHard(hard);         }
+               else                                            {       F.setFilter(string(alignmentLength, '1'));      }
 
                if(trump != '*' || isTrue(vertical) || soft != 0){
                        while(!inFASTA.eof()){  //read through and create the filter...
                                Sequence seq(inFASTA);
-                               if(trump != '*'){       doTrump(seq);   }
-                               if(isTrue(vertical) || soft != 0){      getFreqs(seq);  }
-                               numSeqs++;
-                               cout.flush();
+                               if (seq.getName() != "") {
+                                       if(trump != '*'){       F.doTrump(seq); }
+                                       if(isTrue(vertical) || soft != 0){      F.getFreqs(seq);        }
+                                       numSeqs++;
+                                       cout.flush();
+                               }
                        }
                
                }
                inFASTA.close();
+               F.setNumSeqs(numSeqs);
+               
+               
+               if(isTrue(vertical) == 1)       {       F.doVertical(); }
+               if(soft != 0)                           {       F.doSoft();             }
                
-               if(isTrue(vertical) == 1)       {       doVertical();   }
-               if(soft != 0)   {       doSoft();               }                       
+               filter = F.getFilter();
 
                ofstream outFilter;
-               string filterFile = getRootName(fastafile) + "filter";
+               string filterFile = outputDir + getRootName(getSimpleName(fastafile)) + "filter";
                openOutputFile(filterFile, outFilter);
                outFilter << filter << endl;
                outFilter.close();
                
                ifstream inFasta2;
                openInputFile(fastafile, inFasta2);
-               string filteredFasta = getRootName(fastafile) + "filter.fasta";
+               string filteredFasta = outputDir + getRootName(getSimpleName(fastafile)) + "filter.fasta";
                ofstream outFASTA;
                openOutputFile(filteredFasta, outFASTA);
 
                numSeqs = 0;
                while(!inFasta2.eof()){
                        Sequence seq(inFasta2);
-                       string align = seq.getAligned();
-                       string filterSeq = "";
-       
-                       for(int j=0;j<alignmentLength;j++){
-                               if(filter[j] == '1'){
-                                       filterSeq += align[j];
+                       if (seq.getName() != "") {
+                               string align = seq.getAligned();
+                               string filterSeq = "";
+                               
+                               for(int j=0;j<alignmentLength;j++){
+                                       if(filter[j] == '1'){
+                                               filterSeq += align[j];
+                                       }
                                }
+                               
+                               outFASTA << '>' << seq.getName() << endl << filterSeq << endl;
+                               numSeqs++;
                        }
-
-                       outFASTA << '>' << seq.getName() << endl << filterSeq << endl;
-                       numSeqs++;
                        gobble(inFasta2);
                }
                outFASTA.close();