view commons/tools/BenchmarkTEconsensus.py @ 31:0ab839023fe4

Uploaded

#!/usr/bin/env python

##@file
# Compare two fasta files of TEs to assess how reference sequences are recovered by de novo consensus.


# Copyright INRA (Institut National de la Recherche Agronomique)
# http://www.inra.fr
# http://urgi.versailles.inra.fr
#
# This software is governed by the CeCILL license under French law and
# abiding by the rules of distribution of free software.  You can  use, 
# modify and/ or redistribute the software under the terms of the CeCILL
# license as circulated by CEA, CNRS and INRIA at the following URL
# "http://www.cecill.info". 
#
# As a counterpart to the access to the source code and  rights to copy,
# modify and redistribute granted by the license, users are provided only
# with a limited warranty  and the software's author,  the holder of the
# economic rights,  and the successive licensors  have only  limited
# liability. 
#
# In this respect, the user's attention is drawn to the risks associated
# with loading,  using,  modifying and/or developing or reproducing the
# software by the user in light of its specific status of free software,
# that may mean  that it is complicated to manipulate,  and  that  also
# therefore means  that it is reserved for developers  and  experienced
# professionals having in-depth computer knowledge. Users are therefore
# encouraged to load and test the software's suitability as regards their
# requirements in conditions enabling the security of their systems and/or 
# data to be ensured and,  more generally, to use and operate it in the 
# same conditions as regards security. 
#
# The fact that you are presently reading this means that you have had
# knowledge of the CeCILL license and that you accept its terms.


import os
import sys
import getopt
import shutil
import glob


import pyRepet.launcher.programLauncher
from commons.core.coord.AlignUtils import AlignUtils
from commons.core.coord.MatchUtils import MatchUtils
from commons.core.utils.FileUtils import FileUtils
from commons.core.seq.AlignedBioseqDB import AlignedBioseqDB
from commons.core.seq.FastaUtils import FastaUtils
from commons.tools.ChangeSequenceHeaders import ChangeSequenceHeaders


class BenchmarkTEconsensus( object ):
    
    def __init__( self ):
        self._qryFile = ""
        self._sbjFile = ""
        self._method = 1
        self._keepConflictSbj = False
        self._thresholdCoverage = 95
        self._thresholdIdentity = 80
        self._thresholdEvalue = 1e-10
        self._thresholdCoverageMatch = 90
        self._useCluster = False
        self._queue = ""
        self._configFileName = ""
        self._clean = False
        self._verbose = 0
        self._pL = pyRepet.launcher.programLauncher.programLauncher()
        
        
    def help( self ):
        print
        print "usage: BenchmarkTEconsensus.py [ options ]"
        print "options:"
        print "     -h: this help"
        print "     -q: name of the query file (de novo consensus, format='fasta')"
        print "     -s: name of the subject file (reference sequences, format='fasta')"
        print "     -m: method"
        print "         1: Blaster + Matcher (default)"
        print "         2: Blaster + merge + Matcher (not with '-Q')"
        print "         3: Orienter + Mafft + Matcher"
        print "         4: Yass + Matcher"
        print "     -a: keep all conflicting subjects in Matcher"
        print "     -t: coverage threshold over which the match is 'complete' (in %% of the seq length, default=%i)" % self._thresholdCoverage
        print "     -I: identity threshold for 'CC' matches (default=%i)" % self._thresholdIdentity
        print "     -E: E-value threshold for 'CC' matches (default=1e-10)"
        print "     -T: coverage threshold for match length on query compare to subject length (default=%i)" % self._thresholdCoverageMatch
        print "     -Q: queue name to run in parallel"
        print "     -C: name of the configuration file (compulsory with '-Q')"
        print "     -c: clean"
        print "     -v: verbosity level (default=0/1/2)"
        print
        
        
    def setAttributesFromCmdLine( self ):
        try:
            opts, args = getopt.getopt( sys.argv[1:], "hq:s:m:at:I:E:T:Q:C:cv:" )
        except getopt.GetoptError, err:
            sys.stderr.write( "%s\n" % ( str(err) ) )
            self.help()
            sys.exit(1)
        for o,a in opts:
            if o == "-h":
                self.help();
                sys.exit(0)
            elif o == "-q":
                self._qryFile = a
            elif o == "-s":
                self._sbjFile = a
            elif o == "-m":
                self._method = int(a)
            elif o == "-a":
                self._keepConflictSbj = True
            elif o == "-t":
                self._thresholdCoverage = int(a)
            elif o == "-I":
                self._thresholdIdentity = int(a)
            elif o == "-E":
                self._thresholdEvalue = float(a)
            elif o == "-T":
                self._thresholdCoverageMatch = int(a)
            elif o == "-Q":
                self._useCluster = True
                self._queue = a
            elif o == "-C":
                self._configFile = a
            elif o == "-c":
                self._clean = True
            elif o == "-v":
                self._verbose = int(a)
                
                
    def checkAttributes( self ):
        if self._qryFile == "":
            msg = "ERROR: missing query file (-q)"
            sys.stderr.write( "%s\n" % ( msg ) )
            self.help()
            sys.exit(1)
        if not os.path.exists( self._qryFile ):
            msg = "ERROR: can't find file '%s'" % ( self._qryFile )
            sys.stderr.write( "%s\n" % ( msg ) )
            self.help()
            sys.exit(1)
        if self._sbjFile == "":
            msg = "ERROR: missing subject file (-s)"
            sys.stderr.write( "%s\n" % ( msg ) )
            self.help()
            sys.exit(1)
        if not os.path.exists( self._sbjFile ):
            msg = "ERROR: can't find file '%s'" % ( self._sbjFile )
            sys.stderr.write( "%s\n" % ( msg ) )
            self.help()
            sys.exit(1)
        if self._useCluster:
            if self._configFile == "":
                msg = "ERROR: missing configuration file (-C)"
                sys.stderr.write( "%s\n" % ( msg ) )
                self.help()
                sys.exit(1)
            if not os.path.exists( self._configFile ):
                msg = "ERROR: can't find file '%s'" % ( self._configFile )
                sys.stderr.write( "%s\n" % ( msg ) )
                self.help()
                sys.exit(1)
            if self._method == 2:
                msg = "ERROR: can't launch method 2 in parallel"
                sys.stderr.write( "%s\n" % ( msg ) )
                self.help()
                sys.exit(1)
        nbSeqQry = FastaUtils.dbSize( self._qryFile )
        if nbSeqQry == 0:
            print "WARNING: query file is empty"
            sys.exit(0)
        nbSeqSbj = FastaUtils.dbSize( self._sbjFile )
        if nbSeqSbj == 0:
            print "WARNING: subject file is empty"
            sys.exit(0)
            
            
    def preprocess( self ):
        tmpDir = "tmp%s_t%i_m%i_I%i" % ( os.getpid(),
                                         self._thresholdCoverage,
                                         self._method,
                                         self._thresholdIdentity )
        if os.path.exists( tmpDir ):
            shutil.rmtree( tmpDir )
        os.mkdir( tmpDir )
        os.chdir( tmpDir )
        
        os.symlink( "../%s" % self._qryFile, self._qryFile )
        csh = ChangeSequenceHeaders()
        csh.setInputFile( self._qryFile )
        csh.setFormat( "fasta" )
        csh.setStep( 1 )
        csh.setPrefix( "query" )
        csh.setOutputFile( "%s.newH" % ( self._qryFile ) )
        csh.setVerbosityLevel( self._verbose )
        csh.run()
        self._qryFile += ".newH"
        
        if not os.path.exists( self._sbjFile ):
            os.symlink( "../%s" % self._sbjFile, self._sbjFile )
            
            
    def compareFastaViaBlasterMatcher( self, merged=False ):
        """
        Blaster (+ merged) +  Matcher
        """
        if self._keepConflictSbj:
            s = "match"
        else:
            s = "clean_match"
        matchFile = "%s_vs_%s.m%i.align" % ( self._qryFile,
                                             self._sbjFile,
                                             self._method )
        pathFile = "%s_vs_%s.m%i.align" % ( self._qryFile,
                                            self._sbjFile,
                                            self._method )
        if merged:
            matchFile += ".merged"
            pathFile += ".merged"
        matchFile += ".%s.tab" % ( s )
        pathFile += ".%s.path" % ( s )
        
        if not self._useCluster:
            prg = "blaster"
            cmd = prg
            cmd += " -q %s" % ( self._qryFile )
            cmd += " -s %s" % ( self._sbjFile )
            cmd += " -B %s_vs_%s.m%i" % ( self._qryFile,
                                          self._sbjFile,
                                          self._method )
            cmd += " -v %i" % ( self._verbose )
            self._pL.launch( prg, cmd )
            
            if merged:
                tmpFile = "%s_vs_%s.m%i.align.merged" % ( self._qryFile,
                                                          self._sbjFile,
                                                          self._method )
                AlignUtils.mergeFile( "%s_vs_%s.m%i.align" % ( self._qryFile,
                                                               self._sbjFile,
                                                               self._method ),
                                                               tmpFile )
            else:
                tmpFile = "%s_vs_%s.m%i.align" % ( self._qryFile,
                                                   self._sbjFile,
                                                   self._method )
                
            prg = "matcher"
            cmd = prg
            cmd += " -m %s" % ( tmpFile )
            cmd += " -q %s" % ( self._qryFile )
            cmd += " -s %s" % ( self._sbjFile )
            cmd += " -j"
            if self._keepConflictSbj:
                cmd += " -a"
            cmd += " -v %i" % ( self._verbose )
            self._pL.launch( prg, cmd )
            
        else:
            os.symlink( "../%s" % self._configFile, self._configFileName )
            
            prg = os.environ["REPET_PATH"] + "/bin/launchBlasterMatcherPerQuery.py"
            cmd = prg
            cmd += " -q %s" % ( self._qryFile )
            cmd += " -s %s" % ( self._sbjFile )
            cmd += " -Q %s" % ( self._queue )
            cmd += " -C %s" % ( self._configFile )
            cmd += " -n %i" % ( 10 )
            if self._keepConflictSbj:
                cmd += " -M \"%s\"" % ( "-j -a" )
            else:
                cmd += " -M \"%s\"" % ( "-j" )
            cmd += " -Z tab"
            if self._clean:
                cmd += " -c"
            cmd += " -v %i" % ( self._verbose - 1 )
            self._pL.launch( prg, cmd )
            
        csh = ChangeSequenceHeaders()
        csh.setInputFile( matchFile )
        csh.setFormat( "tab" )
        csh.setStep( 2 )
        csh.setLinkFile( "%slink" % ( self._qryFile ) )
        csh.setOutputFile( matchFile.replace(".newH","") )
        csh.run()
        
        csh.setInputFile( pathFile )
        csh.setFormat( "path" )
        csh.setStep( 2 )
        csh.setOutputFile( pathFile.replace(".newH","") )
        csh.run()
        
        return matchFile.replace(".newH",""), pathFile.replace(".newH","")
    
    
    def compareFastaViaMafft( self ):
        """
        Orienter, Mafft, Matcher
        """
        FastaUtils.dbSplit( self._qryFile, 1, False, False, "query" )
        FastaUtils.dbSplit( self._sbjFile, 1, False, False, "subject" )
        lQueries = glob.glob( "query_*.fa" )
        lSubjects = glob.glob( "subject_*.fa" )
        
        if self._keepConflictSbj:
            s = "match"
        else:
            s = "clean_match"
        matchFile = "%s_vs_%s.m%i.align.%s.tab" % ( self._qryFile,
                                                    self._sbjFile,
                                                    self._method,
                                                    s )
        os.system( "touch %s" %  matchFile )
        pathFile = "%s_vs_%s.m%i.align.%s.path" % ( self._qryFile,
                                                    self._sbjFile,
                                                    self._method,
                                                    s )
        os.system( "touch %s" %  pathFile )
        
        countQueries = 0
        for query in lQueries:
            countQueries += 1
            queryHeader = FastaUtils.dbHeaders( query )[0]
            queryLength = FastaUtils.dbLengths( query )[0]
            countSubjects = 0
            for subject in lSubjects:
                countSubjects += 1
                subjectHeader = FastaUtils.dbHeaders( subject )[0]
                subjectLength = FastaUtils.dbLengths( subject )[0]
                if self._verbose > 0:
                    print "compare '%s' (%i bp, %i/%i) and '%s' (%i bp, %i/%i)" % ( queryHeader, queryLength,
                                                                                    countQueries, len(lQueries),
                                                                                    subjectHeader, subjectLength,
                                                                                    countSubjects, len(lSubjects) )
                    sys.stdout.flush()
                qsLengthRatio = 100 * queryLength / float(subjectLength)
                if qsLengthRatio < self._thresholdCoverage - 2 or qsLengthRatio > 100 + (100-self._thresholdCoverage) + 2:
                    if self._verbose > 0:
                        print "skip (q/s=%.2f%%)" % ( qsLengthRatio )
                    continue
                
                tmpFile = "%s_vs_%s" % ( query, subject )
                FileUtils.catFilesFromList( [ query, subject ],
                                            tmpFile )
                prg = "OrientSequences.py"
                cmd = prg
                cmd += " -i %s" % ( tmpFile )
                cmd += " -p mummer"
                cmd += " -c"
                cmd += " -v %i" % ( self._verbose - 1 )
                self._pL.launch( prg, cmd )
                prg = "MafftProgramLauncher.py"
                cmd = prg
                cmd += " -i %s.oriented" % ( tmpFile )
                cmd += " -c"
                cmd += " -v %i" % ( self._verbose - 1 )
                self._pL.launch( prg, cmd )
                
                absDB = AlignedBioseqDB( "%s.oriented.fa_aln" % tmpFile )
                lHeaders = absDB.getHeaderList()
                lAligns = absDB.getAlignList( lHeaders[0], lHeaders[1] )
                for i in lAligns:
                    if "re-oriented" in i.getQueryName():
                        i.setQueryName( queryHeader )
                        start = i.getQueryStart()
                        end = i.getQueryEnd()
                        i.setQueryStart( queryLength - end + 1 )
                        i.setQueryEnd( queryLength - start + 1 )
                    if "re-oriented" in i.getSubjectName():
                        i.setSubjectName( subjectHeader )
                        start = i.getSubjectStart()
                        end = i.getSubjectEnd()
                        i.setSubjectEnd( subjectLength - end + 1 )
                        i.setSubjectStart( subjectLength - start + 1 )
                    if not i.isQueryOnDirectStrand():
                        i.reverse()
                AlignUtils.writeListInFile( lAligns,
                                            "%s.oriented.fa_aln.align" % tmpFile )
                
                prg = os.environ["REPET_PATH"] + "/bin/matcher"
                cmd = prg
                cmd += " -m %s.oriented.fa_aln.align" % ( tmpFile )
                cmd += " -q %s" % ( query )
                cmd += " -s %s" % ( subject )
                cmd += " -j"
                if self._keepConflictSbj:
                    cmd += " -a"
                cmd += " -v %i" % ( self._verbose - 1 )
                self._pL.launch( prg, cmd )
                
                FileUtils.appendFileContent( "%s.oriented.fa_aln.align.%s.path" % ( tmpFile, s ), pathFile )
                lMatches = MatchUtils.getMatchListFromFile( "%s.oriented.fa_aln.align.%s.tab" % ( tmpFile, s ) )
                MatchUtils.writeListInFile( lMatches, matchFile, "a" )
                
                for f in [ tmpFile,
                          "%s.oriented" % ( tmpFile ),
#                          "%s.oriented.fa_aln" % ( tmpFile ),
                          "%s.oriented.fa_aln.align" % ( tmpFile ),
                          "%s.oriented.fa_aln.align.match.fa" % ( tmpFile ),
                          "%s.oriented.fa_aln.align.match.map" % ( tmpFile ),
                          "%s.oriented.fa_aln.align.match.param" % ( tmpFile ),
                          "%s.oriented.fa_aln.align.match.path" % ( tmpFile ),
                          "%s.oriented.fa_aln.align.match.tab" % ( tmpFile ),
                          ]:
                    os.remove( f )
                    
        if not FileUtils.isEmpty( matchFile ):
            csh = ChangeSequenceHeaders()
            csh.setInputFile( matchFile )
            csh.setFormat( "tab" )
            csh.setStep( 2 )
            csh.setLinkFile( "%slink" % ( self._qryFile ) )
            csh.setOutputFile( matchFile.replace(".newH","") )
            csh.run()
            csh.setInputFile( pathFile )
            csh.setFormat( "path" )
            csh.setStep( 2 )
            csh.setOutputFile( pathFile.replace(".newH","") )
            csh.run()
            return matchFile.replace(".newH",""), pathFile.replace(".newH","")
        else:
            return "", ""
        
        
    def compareFastaViaYassMatcher( self, merged=False ):
        """
        Yass +  Matcher
        """
        if self._keepConflictSbj:
            s = "match"
        else:
            s = "clean_match"
        matchFile = "%s_vs_%s.m%i.align" % ( self._qryFile,
                                             self._sbjFile,
                                             self._method )
        pathFile = "%s_vs_%s.m%i.align" % ( self._qryFile,
                                            self._sbjFile,
                                            self._method )
        if merged:
            matchFile += ".merged"
            pathFile += ".merged"
        matchFile += ".%s.tab" % ( s )
        pathFile += ".%s.path" % ( s )
        
        if not self._useCluster:
            prg = os.environ["REPET_PATH"] + "/bin/YassProgramLauncher.py"
            cmd = prg
            cmd += " -i %s" % ( self._qryFile )
            cmd += " -s %s" % ( self._sbjFile )
            cmd += " -c"
#            cmd += " -p '-i 12'"
            cmd += " -o %s_vs_%s.m%i.align" % ( self._qryFile,
                                                self._sbjFile,
                                                self._method )
            cmd += " -v %i" % ( self._verbose )
            self._pL.launch( prg, cmd )
            
            if merged:
                tmpFile = "%s_vs_%s.m%i.align.merged" % ( self._qryFile,
                                                          self._sbjFile,
                                                          self._method )
                AlignUtils.mergeFile( "%s_vs_%s.m%i.align" % ( self._qryFile,
                                                               self._sbjFile,
                                                               self._method ),
                                                               tmpFile )
            else:
                tmpFile = "%s_vs_%s.m%i.align" % ( self._qryFile,
                                                   self._sbjFile,
                                                   self._method )
                
            prg = os.environ["REPET_PATH"] + "/bin/matcher"
            cmd = prg
            cmd += " -m %s" % ( tmpFile )
            cmd += " -q %s" % ( self._qryFile )
            cmd += " -s %s" % ( self._sbjFile )
            cmd += " -j"
            if self._keepConflictSbj:
                cmd += " -a"
            cmd += " -v %i" % ( self._verbose )
            self._pL.launch( prg, cmd )
            
        csh = ChangeSequenceHeaders()
        csh.setInputFile( matchFile )
        csh.setFormat( "tab" )
        csh.setStep( 2 )
        csh.setLinkFile( "%slink" % ( self._qryFile ) )
        csh.setOutputFile( matchFile.replace(".newH","") )
        csh.run()
        
        csh.setInputFile( pathFile )
        csh.setFormat( "path" )
        csh.setStep( 2 )
        csh.setOutputFile( pathFile.replace(".newH","") )
        csh.run()
        
        return matchFile.replace(".newH",""), pathFile.replace(".newH","")
    
    
    def analyzeMatchFile( self, matchFile, pathFile ):
        if matchFile != "":
            if self._verbose > 0:
                print "analyze the 'tab' file..."
                sys.stdout.flush()
            prg = os.environ["REPET_PATH"] + "/bin/tabFileReader.py"
            cmd = prg
            cmd += " -m %s" % ( matchFile )
            cmd += " -q %s" % ( self._qryFile.replace(".newH","") )
            cmd += " -s %s" % ( self._sbjFile.replace(".newH","") )
            cmd += " -t %i" % ( self._thresholdCoverage )
            cmd += " -I %i" % ( self._thresholdIdentity )
            cmd += " -E %g" % ( self._thresholdEvalue )
            cmd += " -T %i" % ( self._thresholdCoverageMatch )
            cmd += " -v %i" % ( self._verbose - 1 )
            self._pL.launch( prg, cmd )
            for f in [ matchFile, pathFile,
                      "%s_tabFileReader.txt" %  matchFile,
                      "%s_qryCategories.txt" %  matchFile,
                      "%s_sbjCategories.txt" %  matchFile ]:
                shutil.copy( f, ".." )
        os.chdir( ".." )
        
        
    def start( self ):
        self.checkAttributes()
        if self._verbose > 0:
            print "START BenchmarkTEconsensus.py"
            sys.stdout.flush()
            
            
    def end( self ):
        if self._clean:
            tmpDir = "tmp%s_t%i_m%i_I%i" % ( os.getpid(),
                                             self._thresholdCoverage,
                                             self._method,
                                             self._thresholdIdentity )
            shutil.rmtree( tmpDir )
        if self._verbose > 0:
            print "END BenchmarkTEconsensus.py"
            sys.stdout.flush()
            
            
    def run( self ):
        self.start()
        
        self.preprocess()
        
        if self._method == 1:
            matchFile, pathFile = self.compareFastaViaBlasterMatcher()
        elif self._method == 2:
            matchFile, pathFile = self.compareFastaViaBlasterMatcher( merged=True )
        elif self._method == 3:
            matchFile, pathFile = self.compareFastaViaMafft()
        elif self._method == 4:
            matchFile, pathFile = self.compareFastaViaYassMatcher()
            
        self.analyzeMatchFile( matchFile, pathFile )
        
        self.end()
        
        
if __name__ == "__main__":
    i = BenchmarkTEconsensus()
    i.setAttributesFromCmdLine()
    i.run()