Beispiel #1
0
    def getRunStatement(self, infile, outfile, controlfile):
        """
        Generate a specific run statement for each peakcaller class
        """
        # select location of the spp script to run
        if self.PARAMS_PEAKCALLER["spp_options_idr_script"] == "default":
            executable = P.which("run_spp.R")
        elif self.PARAMS_PEAKCALLER["spp_options_idr_script"] == "nodups":
            executable = P.which("run_spp_nodups.R")
        else:
            executable = self.PARAMS_PEAKCALLER["spp_options_idr_script"]
            try:
                os.path.exists(executable)
            except:
                raise IOError("SPP script not found: %s" % executable)

        # select the threshold for lax peak calling
        if self.PARAMS_PEAKCALLER["spp_options_npeaks"]:
            if self.PARAMS_PEAKCALLER["spp_options_fdr"]:
                raise Exception("Value specified for both SPP options"
                                " -npeaks and -fdr please select one or"
                                " other option, but not both")
            else:
                threshold = "-npeaks=" + \
                    str(self.PARAMS_PEAKCALLER["spp_options_npeaks"])
        elif self.PARAMS_PEAKCALLER["spp_options_fdr"]:
            threshold = "-fdr=" + \
                str(self.PARAMS_PEAKCALLER["spp_options_fdr"])
        else:
            raise Exception("Must specify a value for either"
                            " spp_options_npeaks or spp_options_fdr,"
                            " but not both")

        # build run statement for spp.
        # -savn is output.npeak.file (passed as NULL,
        #                             means filename based on infile)
        # -out is output.result.file
        # -odir defaults to os.path.dirname( infile )
        # -savn is save narrowpeak file
        # -savr is save regionpeak file
        #  (run_spp.R script throws an error if region peak is not output).
        statement = [("Rscript %(executable)s"
                      " -c=%(infile)s"
                      " -i=%(controlfile)s"
                      " %(threshold)s"
                      " -savn"
                      " -savr")]

        # add additional options
        statement.append(self.PARAMS_PEAKCALLER["spp_options_parameters"])

        # specify outfile
        statement.append(" -rf"
                         " -out=/stats/phantomPeakStatsReps.tab"
                         " >& %(outfile)s")

        statement = (" ".join(statement) % locals())

        return statement
Beispiel #2
0
    def getRunStatement(self, infile, outfile, controlfile):
        """
        Generate a specific run statement for each peakcaller class
        """
        # select location of the spp script to run
        if self.PARAMS_PEAKCALLER["spp_options_idr_script"] == "default":
            executable = P.which("run_spp.R")
        elif self.PARAMS_PEAKCALLER["spp_options_idr_script"] == "nodups":
            executable = P.which("run_spp_nodups.R")
        else:
            executable = self.PARAMS_PEAKCALLER["spp_options_idr_script"]
            try:
                os.path.exists(executable)
            except:
                raise IOError("SPP script not found: %s" % executable)

        # select the threshold for lax peak calling
        if self.PARAMS_PEAKCALLER["spp_options_npeaks"]:
            if self.PARAMS_PEAKCALLER["spp_options_fdr"]:
                raise Exception("Value specified for both SPP options"
                                " -npeaks and -fdr please select one or"
                                " other option, but not both")
            else:
                threshold = "-npeaks=" + \
                    str(self.PARAMS_PEAKCALLER["spp_options_npeaks"])
        elif self.PARAMS_PEAKCALLER["spp_options_fdr"]:
            threshold = "-fdr=" + \
                str(self.PARAMS_PEAKCALLER["spp_options_fdr"])
        else:
            raise Exception("Must specify a value for either"
                            " spp_options_npeaks or spp_options_fdr,"
                            " but not both")

        # build run statement for spp.
        # -savn is output.npeak.file (passed as NULL,
        #                             means filename based on infile)
        # -out is output.result.file
        # -odir defaults to os.path.dirname( infile )
        # -savn is save narrowpeak file
        # -savr is save regionpeak file
        #  (run_spp.R script throws an error if region peak is not output).
        statement = [("Rscript %(executable)s"
                      " -c=%(infile)s"
                      " -i=%(controlfile)s"
                      " %(threshold)s"
                      " -savn"
                      " -savr")]

        # add additional options
        statement.append(self.PARAMS_PEAKCALLER["spp_options_parameters"])

        # specify outfile
        statement.append(" -rf"
                         " -out=/stats/phantomPeakStatsReps.tab"
                         " >& %(outfile)s")

        statement = (" ".join(statement) % locals())

        return statement
def runControlCPC(infile, outfile):
    # farm.py is called from within cpc.sh
    assert P.which("farm.py"), "farm.py needs to be in $PATH for cpc to run"
    # Default cpc parameters don't work with later versions of blast
    E.info("Running cpc with blast version:%s" % P.which("blastx"))

    result_evidence = P.snip(outfile, ".result") + ".evidence"
    working_dir = "lncRNA_control/cpc"
    statement = ("%(scriptsdir)s/cpc.sh"
                 " %(infile)s"
                 " %(outfile)s"
                 " %(working_dir)s"
                 " %(result_evidence)s")
    P.run()
def runControlCPC(infile, outfile):
    # farm.py is called from within cpc.sh
    assert P.which("farm.py"), "farm.py needs to be in $PATH for cpc to run"
    # Default cpc parameters don't work with later versions of blast
    E.info("Running cpc with blast version:%s" % P.which("blastx"))

    result_evidence = P.snip(outfile, ".result") + ".evidence"
    working_dir = "lncRNA_control/cpc"
    statement = ("%(scriptsdir)s/cpc.sh"
                 " %(infile)s"
                 " %(outfile)s"
                 " %(working_dir)s"
                 " %(result_evidence)s")
    P.run()