* Mothur
*
* Created by Thomas Ryabin on 5/4/09.
- * Copyright 2009 __MyCompanyName__. All rights reserved.
+ * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
*
*/
#include "filterseqscommand.h"
-#include <iostream>
-#include <fstream>
+#include "sequence.hpp"
/**************************************************************************************/
-void FilterSeqsCommand::doTrump() {
- trump = globaldata->getTrump();
- for(int i = 0; i < db->size(); i++) {
- Sequence cur = db->get(i);
- string curAligned = cur.getAligned();
- for(int j = 0; j < curAligned.length(); j++) {
- string curChar = curAligned.substr(j, 1);
- if(curChar.compare(trump) == 0)
- columnsToRemove[j] = true;
- }
- }
-}
-/**************************************************************************************/
-void FilterSeqsCommand::doSoft() {
- soft = atoi(globaldata->getSoft().c_str());
- vector<vector<int> > columnSymbolSums;
- vector<vector<string> > columnSymbols;
- for(int i = 0; i < db->get(0).getLength(); i++) {
- vector<string> symbols;
- vector<int> sums;
- columnSymbols.push_back(symbols);
- columnSymbolSums.push_back(sums);
- }
-
- for(int i = 0; i < db->size(); i++) {
- Sequence cur = db->get(i);
- string curAligned = cur.getAligned();
-
- for(int j = 0; j < curAligned.length(); j++) {
- string curChar = curAligned.substr(j, 1);
- vector<string> curColumnSymbols = columnSymbols[j];
- bool newSymbol = true;
+FilterSeqsCommand::FilterSeqsCommand(string option){
+ try {
+ abort = false;
+
+ //allow user to run help
+ if(option == "help") { help(); abort = true; }
+
+ else {
+ //valid paramters for this command
+ 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();
- for(int k = 0; k < curColumnSymbols.size(); k++)
- if(curChar.compare(curColumnSymbols[k]) == 0) {
- newSymbol = false;
- columnSymbolSums[j][k]++;
+ ValidParameters validParameter;
+ map<string,string>::iterator it;
+
+ //check to make sure all parameters are valid for command
+ 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(newSymbol) {
- columnSymbols[j].push_back(curChar);
- columnSymbolSums[j].push_back(1);
+ //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...
+
+ string temp;
+ temp = validParameter.validFile(parameters, "trump", false); if (temp == "not found") { temp = "*"; }
+ trump = temp[0];
+
+ temp = validParameter.validFile(parameters, "soft", false); if (temp == "not found") { soft = 0; }
+ else { soft = (float)atoi(temp.c_str()) / 100.0; }
+
+ hard = validParameter.validFile(parameters, "hard", true); if (hard == "not found") { hard = ""; }
+ else if (hard == "not open") { abort = true; }
+
+ vertical = validParameter.validFile(parameters, "vertical", false); if (vertical == "not found") { vertical = "T"; }
+
+ numSeqs = 0;
+
+ if (abort == false) {
+
+ if (soft != 0) { F.setSoft(soft); }
+ if (trump != '*') { F.setTrump(trump); }
+
+ }
+
}
- }
-
-
- for(int i = 0; i < columnSymbolSums.size(); i++) {
- int totalSum = 0;
- int max = 0;
- vector<int> curColumnSymbols = columnSymbolSums[i];
-
- for(int j = 0; j < curColumnSymbols.size(); j++) {
- int curSum = curColumnSymbols[j];
- //cout << columnSymbols[i][j] << ": " << curSum << "\n";
- if(curSum > max)
- max = curSum;
- totalSum += curSum;
- }
- //cout << "\n";
- if((double)max/(double)totalSum * 100 < soft)
- columnsToRemove[i] = true;
+ }
+ catch(exception& e) {
+ errorOut(e, "FilterSeqsCommand", "FilterSeqsCommand");
+ exit(1);
}
}
-/**************************************************************************************/
-void FilterSeqsCommand::doFilter() {
- filter = globaldata->getFilter();
- ifstream filehandle;
- openInputFile(filter, filehandle);
-
- char c;
- int count = 0;
- while(!filehandle.eof()) {
- c = filehandle.get();
- if(c == '0')
- columnsToRemove[count] = true;
- count++;
+//**********************************************************************************************************************
+
+void FilterSeqsCommand::help(){
+ try {
+ mothurOut("The filter.seqs command reads a file containing sequences and creates a .filter and .filter.fasta file.\n");
+ mothurOut("The filter.seqs command parameters are fasta, trump, soft, hard and vertical. \n");
+ mothurOut("The fasta parameter is required.\n");
+ mothurOut("The trump parameter .... The default is ...\n");
+ mothurOut("The soft parameter .... The default is ....\n");
+ mothurOut("The hard parameter .... The default is ....\n");
+ mothurOut("The vertical parameter .... The default is T.\n");
+ mothurOut("The filter.seqs command should be in the following format: \n");
+ mothurOut("filter.seqs(fasta=yourFastaFile, trump=yourTrump, soft=yourSoft, hard=yourHard, vertical=yourVertical) \n");
+ mothurOut("Example filter.seqs(fasta=abrecovery.fasta, trump=..., soft=..., hard=..., vertical=T).\n");
+ mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n");
+
+ }
+ catch(exception& e) {
+ errorOut(e, "FilterSeqsCommand", "help");
+ exit(1);
}
}
/**************************************************************************************/
+
int FilterSeqsCommand::execute() {
try {
- globaldata = GlobalData::getInstance();
- db = globaldata->gSequenceDB;
+
+ if (abort == true) { return 0; }
+
+ ifstream inFASTA;
+ openInputFile(fastafile, inFASTA);
+
+ Sequence testSeq(inFASTA);
+ alignmentLength = testSeq.getAlignLength();
+ inFASTA.seekg(0);
+
+ F.setLength(alignmentLength);
+
+ if(soft != 0 || isTrue(vertical)){
+ F.initialize();
+ }
+
+ 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 (seq.getName() != "") {
+ if(trump != '*'){ F.doTrump(seq); }
+ if(isTrue(vertical) || soft != 0){ F.getFreqs(seq); }
+ numSeqs++;
+ cout.flush();
+ }
+ }
+
+ }
+ inFASTA.close();
+ F.setNumSeqs(numSeqs);
- for(int i = 0; i < db->get(0).getLength(); i++)
- columnsToRemove.push_back(false);
+ if(isTrue(vertical) == 1) { F.doVertical(); }
+ if(soft != 0) { F.doSoft(); }
+
+ filter = F.getFilter();
+
+ ofstream outFilter;
+ string filterFile = outputDir + getRootName(getSimpleName(fastafile)) + "filter";
+ openOutputFile(filterFile, outFilter);
+ outFilter << filter << endl;
+ outFilter.close();
+
+ ifstream inFasta2;
+ openInputFile(fastafile, inFasta2);
+ string filteredFasta = outputDir + getRootName(getSimpleName(fastafile)) + "filter.fasta";
+ ofstream outFASTA;
+ openOutputFile(filteredFasta, outFASTA);
+
+ numSeqs = 0;
+ while(!inFasta2.eof()){
+ Sequence seq(inFasta2);
+ if (seq.getName() != "") {
+ string align = seq.getAligned();
+ string filterSeq = "";
- if(globaldata->getTrump().compare("") != 0)
- doTrump();
- else if(globaldata->getSoft().compare("") != 0)
- doSoft();
-
- else if(globaldata->getFilter().compare("") != 0)
- doFilter();
-
- //for(int i = 0; i < columnsToRemove.size(); i++)
-// {
-// cout << "Remove Column " << i << " = ";
-// if(columnsToRemove[i])
-// cout << "true\n";
-// else
-// cout << "false\n";
-// }
- //Creating the new SequenceDB
- SequenceDB newDB;
- for(int i = 0; i < db->size(); i++) {
- Sequence curSeq = db->get(i);
- string curAligned = curSeq.getAligned();
- string curName = curSeq.getName();
- string newAligned = "";
- for(int j = 0; j < curAligned.length(); j++)
- if(!columnsToRemove[j])
- newAligned += curAligned.substr(j, 1);
-
- Sequence newSeq(curName, newAligned);
- newDB.add(newSeq);
+ for(int j=0;j<alignmentLength;j++){
+ if(filter[j] == '1'){
+ filterSeq += align[j];
+ }
+ }
+
+ outFASTA << '>' << seq.getName() << endl << filterSeq << endl;
+ numSeqs++;
+ }
+ gobble(inFasta2);
}
+ outFASTA.close();
+ inFasta2.close();
+
+
+ int filteredLength = 0;
+ for(int i=0;i<alignmentLength;i++){
+ if(filter[i] == '1'){ filteredLength++; }
+ }
+
+ mothurOutEndLine();
+ mothurOut("Length of filtered alignment: " + toString(filteredLength)); mothurOutEndLine();
+ mothurOut("Number of columns removed: " + toString((alignmentLength-filteredLength))); mothurOutEndLine();
+ mothurOut("Length of the original alignment: " + toString(alignmentLength)); mothurOutEndLine();
+ mothurOut("Number of sequences used to construct filter: " + toString(numSeqs)); mothurOutEndLine();
- ofstream outfile;
- outfile.open("filtertest.txt");
- newDB.print(outfile);
- outfile.close();
-
return 0;
+
}
catch(exception& e) {
- cout << "Standard Error: " << e.what() << " has occurred in the FilterSeqsCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
- exit(1);
- }
- catch(...) {
- cout << "An unknown error has occurred in the FilterSeqsCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ errorOut(e, "FilterSeqsCommand", "execute");
exit(1);
}
}
+
/**************************************************************************************/