5 * Created by Sarah Westcott on 5/15/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
8 * This version of nast does everything I think that the greengenes nast server does and then some. I have added the
9 * feature of allowing users to define their database, kmer size for searching, alignment penalty values and alignment
10 * method. This latter feature is perhaps most significant. nastPlus enables a user to use either a Needleman-Wunsch
11 * (non-affine gap penalty) or Gotoh (affine gap penalty) pairwise alignment algorithm. This is significant because it
12 * allows for a global alignment and not the local alignment provided by bLAst. Furthermore, it has the potential to
13 * provide a better alignment because of the banding method employed by blast (I'm not sure about this).
17 #include "aligncommand.h"
18 #include "sequence.hpp"
20 #include "gotohoverlap.hpp"
21 #include "needlemanoverlap.hpp"
22 #include "blastalign.hpp"
23 #include "noalign.hpp"
26 #include "suffixdb.hpp"
27 #include "blastdb.hpp"
30 #include "nastreport.hpp"
33 //**********************************************************************************************************************
35 AlignCommand::AlignCommand(string option){
37 // globaldata = GlobalData::getInstance();
40 //allow user to run help
41 if(option == "help") { help(); abort = true; }
45 //valid paramters for this command
46 string AlignArray[] = {"template","candidate","search","ksize","align","match","mismatch","gapopen","gapextend", "processors"};
47 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
49 OptionParser parser(option);
50 map<string, string> parameters = parser.getParameters();
52 ValidParameters validParameter;
54 //check to make sure all parameters are valid for command
55 for (map<string, string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
56 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
59 //check for required parameters
60 templateFileName = validParameter.validFile(parameters, "template", true);
61 if (templateFileName == "not found") { cout << "template is a required parameter for the align.seqs command." << endl; abort = true; }
62 else if (templateFileName == "not open") { abort = true; }
64 candidateFileName = validParameter.validFile(parameters, "candidate", true);
65 if (candidateFileName == "not found") { cout << "candidate is a required parameter for the align.seqs command." << endl; abort = true; }
66 else if (candidateFileName == "not open") { abort = true; }
68 //check for optional parameter and set defaults
69 // ...at some point should added some additional type checking...
71 temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found"){ temp = "8"; }
72 convert(temp, kmerSize);
74 temp = validParameter.validFile(parameters, "match", false); if (temp == "not found"){ temp = "1.0"; }
77 temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found"){ temp = "-1.0"; }
78 convert(temp, misMatch);
80 temp = validParameter.validFile(parameters, "gapopen", false); if (temp == "not found"){ temp = "-2.0"; }
81 convert(temp, gapOpen);
83 temp = validParameter.validFile(parameters, "gapextend", false); if (temp == "not found"){ temp = "-1.0"; }
84 convert(temp, gapExtend);
86 temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; }
87 convert(temp, processors);
89 search = validParameter.validFile(parameters, "search", false); if (search == "not found"){ search = "kmer"; }
91 align = validParameter.validFile(parameters, "align", false); if (align == "not found"){ align = "needleman"; }
96 cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function AlignCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
100 cout << "An unknown error has occurred in the AlignCommand class function AlignCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
105 //**********************************************************************************************************************
107 AlignCommand::~AlignCommand(){
112 //**********************************************************************************************************************
114 void AlignCommand::help(){
116 cout << "The align.seqs command reads a file containing sequences and creates an alignment file and a report file." << "\n";
117 cout << "The align.seqs command parameters are template, candidate, search, ksize, align, match, mismatch, gapopen and gapextend. " << "\n";
118 cout << "The template and candidate parameters are required." << "\n";
119 cout << "The search parameter allows you to specify the method to find most similar template. Your options are: suffix, kmer and blast. The default is kmer." << "\n";
120 cout << "The align parameter allows you to specify the alignment method to use. Your options are: gotoh, needleman, blast and noalign. The default is needleman." << "\n";
121 cout << "The ksize parameter allows you to specify the kmer size for finding most similar template to candidate. The default is 7." << "\n";
122 cout << "The match parameter allows you to specify the bonus for having the same base. The default is 1.0." << "\n";
123 cout << "The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0." << "\n";
124 cout << "The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -1.0." << "\n";
125 cout << "The gapextend parameter allows you to specify the penalty for extending a gap in an alignment. The default is -2.0." << "\n";
126 cout << "The align.seqs command should be in the following format: " << "\n";
127 cout << "align.seqs(template=yourTemplateFile, candidate=yourCandidateFile, align=yourAlignmentMethod, search=yourSearchmethod, ksize=yourKmerSize, match=yourMatchBonus, mismatch=yourMismatchpenalty, gapopen=yourGapopenPenalty, gapextend=yourGapExtendPenalty) " << "\n";
128 cout << "Example align.seqs(candidate=candidate.fasta, template=core.filtered, align=kmer, search=gotoh, ksize=8, match=2.0, mismatch=3.0, gapopen=-2.0, gapextend=-1.0)" << "\n";
129 cout << "Note: No spaces between parameter labels (i.e. candidate), '=' and parameters (i.e.yourFastaFile)." << "\n" << "\n";
131 catch(exception& e) {
132 cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
136 cout << "An unknown error has occurred in the AlignCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
142 //**********************************************************************************************************************
144 int AlignCommand::execute(){
146 if (abort == true) { return 0; }
148 srand( (unsigned)time( NULL ) ); //needed to assign names to temporary files
151 if(search == "kmer") { templateDB = new KmerDB(templateFileName, kmerSize); }
152 else if(search == "suffix") { templateDB = new SuffixDB(templateFileName); }
153 else if(search == "blast") { templateDB = new BlastDB(templateFileName, gapOpen, gapExtend, match, misMatch); }
155 cout << search << " is not a valid search option. I will run the command using kmer, ksize=8." << endl; kmerSize = 8;
156 templateDB = new KmerDB(templateFileName, kmerSize);
159 if(align == "gotoh") { alignment = new GotohOverlap(gapOpen, gapExtend, match, misMatch, 3000); }
160 else if(align == "needleman") { alignment = new NeedlemanOverlap(gapOpen, match, misMatch, 3000); }
161 else if(align == "blast") { alignment = new BlastAlignment(gapOpen, gapExtend, match, misMatch); }
162 else if(align == "noalign") { alignment = new NoAlign(); }
164 cout << align << " is not a valid alignment option. I will run the command using needleman." << endl;
165 alignment = new NeedlemanOverlap(gapOpen, match, misMatch, 3000);
168 string alignFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "align";
169 string reportFileName = candidateFileName.substr(0,candidateFileName.find_last_of(".")+1) + "align.report";
171 int numFastaSeqs = 0;
172 int start = time(NULL);
174 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
177 openInputFile(candidateFileName, inFASTA);
178 numFastaSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
181 lines.push_back(new linePair(0, numFastaSeqs));
183 driver(lines[0], alignFileName, reportFileName);
186 vector<int> positions;
187 processIDS.resize(0);
190 openInputFile(candidateFileName, inFASTA);
192 while(!inFASTA.eof()){
193 char c = inFASTA.get();
194 if(c == '>'){ positions.push_back(inFASTA.tellg()); }
195 while (!inFASTA.eof()) { c = inFASTA.get(); if (c == 10 || c == 13){ break; } } // get rest of line if there's any crap there
199 numFastaSeqs = positions.size();
201 int numSeqsPerProcessor = numFastaSeqs / processors;
203 for (int i = 0; i < processors; i++) {
204 int startPos = positions[ i * numSeqsPerProcessor ];
205 if(i == processors - 1){
206 numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
208 lines.push_back(new linePair(startPos, numSeqsPerProcessor));
210 createProcesses(alignFileName, reportFileName);
212 rename((alignFileName + toString(processIDS[0]) + ".temp").c_str(), alignFileName.c_str());
213 rename((reportFileName + toString(processIDS[0]) + ".temp").c_str(), reportFileName.c_str());
215 for(int i=1;i<processors;i++){
216 appendAlignFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName);
217 remove((alignFileName + toString(processIDS[i]) + ".temp").c_str());
219 appendReportFiles((reportFileName + toString(processIDS[i]) + ".temp"), reportFileName);
220 remove((reportFileName + toString(processIDS[i]) + ".temp").c_str());
225 openInputFile(candidateFileName, inFASTA);
226 numFastaSeqs=count(istreambuf_iterator<char>(inFASTA),istreambuf_iterator<char>(), '>');
229 lines.push_back(new linePair(0, numFastaSeqs));
231 driver(lines[0], alignFileName, reportFileName);
234 cout << "It took " << time(NULL) - start << " secs to align " << numFastaSeqs << " sequences" << endl;
241 catch(exception& e) {
242 cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
246 cout << "An unknown error has occurred in the AlignCommand class function execute. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
251 //**********************************************************************************************************************
253 int AlignCommand::driver(linePair* line, string alignFName, string reportFName){
255 ofstream alignmentFile;
256 openOutputFile(alignFName, alignmentFile);
257 NastReport report(reportFName);
260 openInputFile(candidateFileName, inFASTA);
261 inFASTA.seekg(line->start);
263 for(int i=0;i<line->numSeqs;i++){
265 Sequence* candidateSeq = new Sequence(inFASTA);
266 report.setCandidate(candidateSeq);
268 Sequence* templateSeq = templateDB->findClosestSequence(candidateSeq);
269 report.setTemplate(templateSeq);
270 report.setSearchParameters(search, templateDB->getSearchScore());
272 Nast nast(alignment, candidateSeq, templateSeq);
273 report.setAlignmentParameters(align, alignment);
274 report.setNastParameters(nast);
276 alignmentFile << '>' << candidateSeq->getName() << '\n' << candidateSeq->getAligned() << endl;
283 alignmentFile.close();
288 catch(exception& e) {
289 cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function driver. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
293 cout << "An unknown error has occurred in the AlignCommand class function driver. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
298 /**************************************************************************************************/
300 void AlignCommand::createProcesses(string alignFileName, string reportFileName) {
302 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
304 // processIDS.resize(0);
306 //loop through and create all the processes you want
307 while (process != processors) {
311 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
314 driver(lines[process], alignFileName + toString(getpid()) + ".temp", reportFileName + toString(getpid()) + ".temp");
316 }else { cout << "unable to spawn the necessary processes." << endl; exit(0); }
319 //force parent to wait until all the processes are done
320 for (int i=0;i<processors;i++) {
321 int temp = processIDS[i];
326 catch(exception& e) {
327 cout << "Standard Error: " << e.what() << " has occurred in the AlignCommand class Function createProcesses. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
331 cout << "An unknown error has occurred in the AlignCommand class function createProcesses. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
336 /**************************************************************************************************/
338 void AlignCommand::appendAlignFiles(string temp, string filename) {
341 //open output file in append mode
343 openOutputFileAppend(filename, output);
345 //open temp file for reading
347 openInputFile(temp, input);
350 //read input file and write to output file
351 while(input.eof() != true) {
352 getline(input, line); //getline removes the newline char
354 output << line << endl; // Appending back newline char
361 catch(exception& e) {
362 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
366 cout << "An unknown error has occurred in the DistanceCommand class function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
371 /**************************************************************************************************/
373 void AlignCommand::appendReportFiles(string temp, string filename) {
376 //open output file in append mode
378 openOutputFileAppend(filename, output);
380 //open temp file for reading
382 openInputFile(temp, input);
383 while (!input.eof()) { char c = input.get(); if (c == 10 || c == 13){ break; } } // get header line
386 //read input file and write to output file
387 while(input.eof() != true) {
388 getline(input, line); //getline removes the newline char
390 output << line << endl; // Appending back newline char
397 catch(exception& e) {
398 cout << "Standard Error: " << e.what() << " has occurred in the DistanceCommand class Function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
402 cout << "An unknown error has occurred in the DistanceCommand class function appendFiles. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
407 //**********************************************************************************************************************