Exemplo n.º 1
0
    def run(self):
        """Run"""
        iceq = IceQuiver(root_dir=self.root_dir, bas_fofn=self.bas_fofn,
                         fasta_fofn=self.fasta_fofn, sge_opts=self.sge_opts,
                         tmp_dir=self.tmp_dir)
        iceq.validate_inputs()
        iceq.run()

        icepq = IceQuiverPostprocess(root_dir=self.root_dir,
                                     use_sge=self.sge_opts.use_sge,
                                     quit_if_not_done=False,
                                     ipq_opts=self.ipq_opts)
        icepq.run()
        return 0
Exemplo n.º 2
0
    def run(self):
        """Run"""
        iceq = IceQuiver(root_dir=self.root_dir,
                         bas_fofn=self.bas_fofn,
                         fasta_fofn=self.fasta_fofn,
                         sge_opts=self.sge_opts,
                         tmp_dir=self.tmp_dir)
        iceq.validate_inputs()
        iceq.run()

        icepq = IceQuiverPostprocess(root_dir=self.root_dir,
                                     use_sge=self.sge_opts.use_sge,
                                     quit_if_not_done=False,
                                     ipq_opts=self.ipq_opts)
        icepq.run()
        return 0
Exemplo n.º 3
0
 def cluster_dir(self, cid):
     dir_name = IceQuiver.cluster_dir(self, cid)
     if not op.isdir(dir_name):
         os.makedirs(dir_name)
     return dir_name
Exemplo n.º 4
0
    def run(self):
        """Run"""

        for chunk in self.i:
            iceq = IceQuiver(root_dir=self.root_dir,
                             bas_fofn=self.bas_fofn,
                             fasta_fofn=self.fasta_fofn,
                             sge_opts=self.sge_opts,
                             tmp_dir=self.tmp_dir,
                             prog_name="ice_quiver_{i}of{N}".format(i=chunk,
                                                                    N=self.N))
            cmd_str = self.cmd_str()
            iceq.add_log(cmd_str)

            iceq.add_log("root_dir: {d}.".format(d=self.root_dir))
            iceq.add_log("Total number of chunks: N = {N}.".format(N=self.N))
            iceq.add_log("Index of chunks: i = {i}.".format(i=chunk))

            iceq.validate_inputs()

            iceq.process_chunk_i(i=chunk, num_chunks=self.N)
            iceq.close_log()
Exemplo n.º 5
0
    def run(self):
        """
        First, split non-full-length (nfl) fasta files into smaller
        chunks, assign nfl reads in each splitted fasta file
        into unpolished isoform clusters and then merge all pickles
        into self.nfl_all_pickle_fn.
        Second, bin every 100 clusters, for each bin, call blasr,
        samto5h, loadPulses, cmph5tools to create cmp.h5 files and
        call quiver to polish each isoforms within each bin.
        Finally, pick up good isoform clusters whose QV errors is less
        than a threshold.
        Save all high quality isoforms to hq_isoforms_fa|fq if they are not None
        Save all low quality isoforms to lq_isoforms_fa|fq if they are not None
        """
        if guess_file_format(self.bas_fofn) != FILE_FORMATS.BAM:
            # Create input.fasta.fofn from bas_fofn
            self.add_log("Creating fasta fofn from bas/bax.h5/bam fofn",
                         level=logging.INFO)
            if self.fasta_fofn is None:
                self.fasta_fofn = op.join(self.nfl_dir, "input.fasta.fofn")
            self.add_log("fasta fofn={f}".format(f=self.fasta_fofn))
            convert_fofn_to_fasta(fofn_filename=self.bas_fofn,
                                  out_filename=self.fasta_fofn,
                                  fasta_out_dir=self.nfl_dir)
        else:
            self.fasta_fofn = None

        # Split non-full-length reads into smaller fasta files
        # and save files to root_dir/nfl_00.fasta, ..., .
        self.add_log("Splitting {nfl} into ".format(nfl=self.nfl_fa) +
                     "smaller files each containing {n} reads.".format(
                         n=self.ice_opts.nfl_reads_per_split),
                     level=logging.INFO)
        self._nfl_splitted_fas = splitFasta(
            input_fasta=self.nfl_fa,
            reads_per_split=self.ice_opts.nfl_reads_per_split,
            out_dir=self.nfl_dir,
            out_prefix="input.split")
        msg = "Splitted files are: " + "\n".join(self._nfl_splitted_fas)
        self.add_log(msg, level=logging.INFO)

        # Generating dazz DB for final.consensus.fasta
        ref_obj = DazzIDHandler(input_filename=self.final_consensus_fa,
                                converted=False)
        ref_obj.make_db()
        msg = "Dazz DB made for: " + ref_obj.dazz_filename
        self.add_log(msg, level=logging.INFO)

        # Process nfl reads in each splitted fasta.
        self.add_log("Initializing IceAllPartials.", level=logging.INFO)

        self.icep = IceAllPartials(root_dir=self.root_dir,
                                   fasta_filenames=self._nfl_splitted_fas,
                                   ref_fasta=self.final_consensus_fa,
                                   out_pickle=self.nfl_all_pickle_fn,
                                   sge_opts=self.sge_opts,
                                   ccs_fofn=self.ccs_fofn)
        self.add_log("IceAllPartials log: {f}.".format(f=self.icep.log_fn),
                     level=logging.INFO)
        self.icep.run()
        self.add_log("IceAllPartials completed.", level=logging.INFO)

        self.add_log("Initializing IceQuiver.", level=logging.INFO)
        self.iceq = IceQuiver(root_dir=self.root_dir,
                              bas_fofn=self.bas_fofn,
                              fasta_fofn=self.fasta_fofn,
                              sge_opts=self.sge_opts,
                              tmp_dir=self.tmp_dir)
        self.add_log("IceQuiver log: {f}.".format(f=self.iceq.log_fn),
                     level=logging.INFO)
        self.iceq.run()
        self.add_log("IceQuiver finished.", level=logging.INFO)

        self.add_log("Initializing IceQuiverPostprocess.", level=logging.INFO)
        self.icepq = IceQuiverPostprocess(root_dir=self.root_dir,
                                          use_sge=self.sge_opts.use_sge,
                                          quit_if_not_done=False,
                                          ipq_opts=self.ipq_opts)
        self.add_log(
            "IceQuiverPostprocess log: {f}.".format(f=self.icepq.log_fn),
            level=logging.INFO)
        self.icepq.run()
        self.add_log("IceQuiverPostprocess finished.", level=logging.INFO)
Exemplo n.º 6
0
class Polish(IceFiles):
    """Polish isoforms clusters using Quiver."""
    def __init__(self,
                 root_dir,
                 nfl_fa,
                 bas_fofn,
                 ccs_fofn,
                 ice_opts,
                 sge_opts,
                 ipq_opts,
                 fasta_fofn=None,
                 tmp_dir=None):
        """
        root_dir --- IceFiles.root_dir, usually data/clusterOutDir
        nfl_fa    --- non-full-length reads in fasta, e.g., isoseq_nfl.fasta
        bas_fofn --- e.g. input.fofn of bas|bax.h5 files
        ccs_fofn --- e.g. ccs.fofn of ccs files.

        ipq_opts --- IceQuiverHQLQOptions
                     qv_trim_5: ignore QV of n bases in the 5' end
                     qv_trim_3: ignore QV of n bases in the 3' end
                     hq_quiver_min_accuracy: minimum allowed quiver accuracy
                                      to mark an isoform as high quality
                     hq_isoforms_fa|fq: polished, hiqh quality consensus
                                        isoforms in fasta|q
                     lq_isoforms_fa|fq: polished, low quality consensus
                                        isoforms in fasta|q
        """
        IceFiles.__init__(self,
                          prog_name="IcePolish",
                          root_dir=root_dir,
                          bas_fofn=bas_fofn,
                          ccs_fofn=ccs_fofn,
                          fasta_fofn=fasta_fofn,
                          tmp_dir=tmp_dir)
        self.nfl_fa = realpath(nfl_fa)
        self.ice_opts = ice_opts
        self.sge_opts = sge_opts
        self.ipq_opts = ipq_opts

        self.add_log("ece_penalty: {0}, ece_min_len: {1}".format(
            self.ice_opts.ece_penalty, self.ice_opts.ece_min_len))

        self.icep = None  # IceAllPartials.
        self.iceq = None  # IceQuiver
        self.icepq = None  # IceQuiverPostprocess
        self._nfl_splitted_fas = None

        self.validate_inputs()

    def validate_inputs(self):
        """
        Validate input directories: root_dir, and
        files: nfl_fa, bas_fofn, ccs_fofn.
        """
        self.add_log("Validating inputs.")
        errMsg = ""
        if not op.exists(self.root_dir):
            errMsg = "Root dir {d} is not an existing directory!".\
                format(d=self.root_dir)
        if not op.exists(self.nfl_fa):
            errMsg = "Failed to find non-full-length reads {f}!".\
                format(f=self.nfl_fa)
        if self.bas_fofn is None:
            errMsg = "bas_fofn (subreadset) must be specified."
        if not op.exists(self.bas_fofn):
            errMsg = "Failed to find bas fofn (subreadset) {f}!".format(
                f=self.bas_fofn)
        if errMsg != "":
            self.add_log(errMsg, level=logging.ERROR)
            raise ValueError(errMsg)

    def run(self):
        """
        First, split non-full-length (nfl) fasta files into smaller
        chunks, assign nfl reads in each splitted fasta file
        into unpolished isoform clusters and then merge all pickles
        into self.nfl_all_pickle_fn.
        Second, bin every 100 clusters, for each bin, call blasr,
        samto5h, loadPulses, cmph5tools to create cmp.h5 files and
        call quiver to polish each isoforms within each bin.
        Finally, pick up good isoform clusters whose QV errors is less
        than a threshold.
        Save all high quality isoforms to hq_isoforms_fa|fq if they are not None
        Save all low quality isoforms to lq_isoforms_fa|fq if they are not None
        """
        if guess_file_format(self.bas_fofn) != FILE_FORMATS.BAM:
            # Create input.fasta.fofn from bas_fofn
            self.add_log("Creating fasta fofn from bas/bax.h5/bam fofn",
                         level=logging.INFO)
            if self.fasta_fofn is None:
                self.fasta_fofn = op.join(self.nfl_dir, "input.fasta.fofn")
            self.add_log("fasta fofn={f}".format(f=self.fasta_fofn))
            convert_fofn_to_fasta(fofn_filename=self.bas_fofn,
                                  out_filename=self.fasta_fofn,
                                  fasta_out_dir=self.nfl_dir)
        else:
            self.fasta_fofn = None

        # Split non-full-length reads into smaller fasta files
        # and save files to root_dir/nfl_00.fasta, ..., .
        self.add_log("Splitting {nfl} into ".format(nfl=self.nfl_fa) +
                     "smaller files each containing {n} reads.".format(
                         n=self.ice_opts.nfl_reads_per_split),
                     level=logging.INFO)
        self._nfl_splitted_fas = splitFasta(
            input_fasta=self.nfl_fa,
            reads_per_split=self.ice_opts.nfl_reads_per_split,
            out_dir=self.nfl_dir,
            out_prefix="input.split")
        msg = "Splitted files are: " + "\n".join(self._nfl_splitted_fas)
        self.add_log(msg, level=logging.INFO)

        # Generating dazz DB for final.consensus.fasta
        ref_obj = DazzIDHandler(input_filename=self.final_consensus_fa,
                                converted=False)
        ref_obj.make_db()
        msg = "Dazz DB made for: " + ref_obj.dazz_filename
        self.add_log(msg, level=logging.INFO)

        # Process nfl reads in each splitted fasta.
        self.add_log("Initializing IceAllPartials.", level=logging.INFO)

        self.icep = IceAllPartials(root_dir=self.root_dir,
                                   fasta_filenames=self._nfl_splitted_fas,
                                   ref_fasta=self.final_consensus_fa,
                                   out_pickle=self.nfl_all_pickle_fn,
                                   sge_opts=self.sge_opts,
                                   ccs_fofn=self.ccs_fofn)
        self.add_log("IceAllPartials log: {f}.".format(f=self.icep.log_fn),
                     level=logging.INFO)
        self.icep.run()
        self.add_log("IceAllPartials completed.", level=logging.INFO)

        self.add_log("Initializing IceQuiver.", level=logging.INFO)
        self.iceq = IceQuiver(root_dir=self.root_dir,
                              bas_fofn=self.bas_fofn,
                              fasta_fofn=self.fasta_fofn,
                              sge_opts=self.sge_opts,
                              tmp_dir=self.tmp_dir)
        self.add_log("IceQuiver log: {f}.".format(f=self.iceq.log_fn),
                     level=logging.INFO)
        self.iceq.run()
        self.add_log("IceQuiver finished.", level=logging.INFO)

        self.add_log("Initializing IceQuiverPostprocess.", level=logging.INFO)
        self.icepq = IceQuiverPostprocess(root_dir=self.root_dir,
                                          use_sge=self.sge_opts.use_sge,
                                          quit_if_not_done=False,
                                          ipq_opts=self.ipq_opts)
        self.add_log(
            "IceQuiverPostprocess log: {f}.".format(f=self.icepq.log_fn),
            level=logging.INFO)
        self.icepq.run()
        self.add_log("IceQuiverPostprocess finished.", level=logging.INFO)
Exemplo n.º 7
0
    def run(self):
        """Run"""

        for chunk in self.i:
            iceq = IceQuiver(root_dir=self.root_dir,
                             bas_fofn=self.bas_fofn,
                             fasta_fofn=self.fasta_fofn,
                             sge_opts=self.sge_opts,
                             tmp_dir=self.tmp_dir,
                             prog_name="ice_quiver_{i}of{N}".
                                       format(i=chunk, N=self.N))
            cmd_str = self.cmd_str()
            iceq.add_log(cmd_str)

            iceq.add_log("root_dir: {d}.".format(d=self.root_dir))
            iceq.add_log("Total number of chunks: N = {N}.".format(N=self.N))
            iceq.add_log("Index of chunks: i = {i}.".format(i=chunk))

            iceq.validate_inputs()

            iceq.process_chunk_i(i=chunk, num_chunks=self.N)
            iceq.close_log()
Exemplo n.º 8
0
    def run(self):
        """
        First, split non-full-length (nfl) fasta files into smaller
        chunks, assign nfl reads in each splitted fasta file
        into unpolished isoform clusters and then merge all pickles
        into self.nfl_all_pickle_fn.
        Second, bin every 100 clusters, for each bin, call blasr,
        samto5h, loadPulses, cmph5tools to create cmp.h5 files and
        call quiver to polish each isoforms within each bin.
        Finally, pick up good isoform clusters whose QV errors is less
        than a threshold.
        Save all high quality isoforms to hq_isoforms_fa|fq if they are not None
        Save all low quality isoforms to lq_isoforms_fa|fq if they are not None
        """
        if guess_file_format(self.bas_fofn) != FILE_FORMATS.BAM:
            # Create input.fasta.fofn from bas_fofn
            self.add_log("Creating fasta fofn from bas/bax.h5/bam fofn",
                         level=logging.INFO)
            if self.fasta_fofn is None:
                self.fasta_fofn = op.join(self.nfl_dir, "input.fasta.fofn")
            self.add_log("fasta fofn={f}".format(f=self.fasta_fofn))
            convert_fofn_to_fasta(fofn_filename=self.bas_fofn,
                                  out_filename=self.fasta_fofn,
                                  fasta_out_dir=self.nfl_dir)
        else:
            self.fasta_fofn = None

        # Split non-full-length reads into smaller fasta files
        # and save files to root_dir/nfl_00.fasta, ..., .
        self.add_log("Splitting {nfl} into ".format(nfl=self.nfl_fa) +
                     "smaller files each containing {n} reads.".format(
                     n=self.ice_opts.nfl_reads_per_split),
                     level=logging.INFO)
        self._nfl_splitted_fas = splitFasta(input_fasta=self.nfl_fa,
                                            reads_per_split=self.ice_opts.nfl_reads_per_split,
                                            out_dir=self.nfl_dir,
                                            out_prefix="input.split")
        msg = "Splitted files are: " + "\n".join(self._nfl_splitted_fas)
        self.add_log(msg, level=logging.INFO)

        # Generating dazz DB for final.consensus.fasta
        ref_obj = DazzIDHandler(input_filename=self.final_consensus_fa, converted=False)
        ref_obj.make_db()
        msg = "Dazz DB made for: " + ref_obj.dazz_filename
        self.add_log(msg, level=logging.INFO)

        # Process nfl reads in each splitted fasta.
        self.add_log("Initializing IceAllPartials.", level=logging.INFO)

        self.icep = IceAllPartials(
            root_dir=self.root_dir,
            fasta_filenames=self._nfl_splitted_fas,
            ref_fasta=self.final_consensus_fa,
            out_pickle=self.nfl_all_pickle_fn,
            sge_opts=self.sge_opts,
            ccs_fofn=self.ccs_fofn)
        self.add_log("IceAllPartials log: {f}.".format(f=self.icep.log_fn),
                     level=logging.INFO)
        self.icep.run()
        self.add_log("IceAllPartials completed.", level=logging.INFO)

        self.add_log("Initializing IceQuiver.", level=logging.INFO)
        self.iceq = IceQuiver(root_dir=self.root_dir,
                              bas_fofn=self.bas_fofn,
                              fasta_fofn=self.fasta_fofn,
                              sge_opts=self.sge_opts,
                              tmp_dir=self.tmp_dir)
        self.add_log("IceQuiver log: {f}.".format(f=self.iceq.log_fn),
                     level=logging.INFO)
        self.iceq.run()
        self.add_log("IceQuiver finished.", level=logging.INFO)

        self.add_log("Initializing IceQuiverPostprocess.", level=logging.INFO)
        self.icepq = IceQuiverPostprocess(root_dir=self.root_dir,
                                          use_sge=self.sge_opts.use_sge,
                                          quit_if_not_done=False,
                                          ipq_opts=self.ipq_opts)
        self.add_log("IceQuiverPostprocess log: {f}.".
                     format(f=self.icepq.log_fn), level=logging.INFO)
        self.icepq.run()
        self.add_log("IceQuiverPostprocess finished.", level=logging.INFO)
Exemplo n.º 9
0
class Polish(IceFiles):

    """Polish isoforms clusters using Quiver."""

    def __init__(self, root_dir, nfl_fa, bas_fofn, ccs_fofn,
                 ice_opts, sge_opts, ipq_opts, fasta_fofn=None,
                 tmp_dir=None):
        """
        root_dir --- IceFiles.root_dir, usually data/clusterOutDir
        nfl_fa    --- non-full-length reads in fasta, e.g., isoseq_nfl.fasta
        bas_fofn --- e.g. input.fofn of bas|bax.h5 files
        ccs_fofn --- e.g. ccs.fofn of ccs files.

        ipq_opts --- IceQuiverHQLQOptions
                     qv_trim_5: ignore QV of n bases in the 5' end
                     qv_trim_3: ignore QV of n bases in the 3' end
                     hq_quiver_min_accuracy: minimum allowed quiver accuracy
                                      to mark an isoform as high quality
                     hq_isoforms_fa|fq: polished, hiqh quality consensus
                                        isoforms in fasta|q
                     lq_isoforms_fa|fq: polished, low quality consensus
                                        isoforms in fasta|q
        """
        IceFiles.__init__(self, prog_name="IcePolish", root_dir=root_dir,
                          bas_fofn=bas_fofn, ccs_fofn=ccs_fofn,
                          fasta_fofn=fasta_fofn, tmp_dir=tmp_dir)
        self.nfl_fa = realpath(nfl_fa)
        self.ice_opts = ice_opts
        self.sge_opts = sge_opts
        self.ipq_opts = ipq_opts

        self.add_log("ece_penalty: {0}, ece_min_len: {1}".format(self.ice_opts.ece_penalty, self.ice_opts.ece_min_len))

        self.icep = None   # IceAllPartials.
        self.iceq = None   # IceQuiver
        self.icepq = None  # IceQuiverPostprocess
        self._nfl_splitted_fas = None

        self.validate_inputs()

    def validate_inputs(self):
        """
        Validate input directories: root_dir, and
        files: nfl_fa, bas_fofn, ccs_fofn.
        """
        self.add_log("Validating inputs.")
        errMsg = ""
        if not op.exists(self.root_dir):
            errMsg = "Root dir {d} is not an existing directory!".\
                format(d=self.root_dir)
        if not op.exists(self.nfl_fa):
            errMsg = "Failed to find non-full-length reads {f}!".\
                format(f=self.nfl_fa)
        if self.bas_fofn is None:
            errMsg = "bas_fofn (subreadset) must be specified."
        if not op.exists(self.bas_fofn):
            errMsg = "Failed to find bas fofn (subreadset) {f}!".format(f=self.bas_fofn)
        if errMsg != "":
            self.add_log(errMsg, level=logging.ERROR)
            raise ValueError(errMsg)

    def run(self):
        """
        First, split non-full-length (nfl) fasta files into smaller
        chunks, assign nfl reads in each splitted fasta file
        into unpolished isoform clusters and then merge all pickles
        into self.nfl_all_pickle_fn.
        Second, bin every 100 clusters, for each bin, call blasr,
        samto5h, loadPulses, cmph5tools to create cmp.h5 files and
        call quiver to polish each isoforms within each bin.
        Finally, pick up good isoform clusters whose QV errors is less
        than a threshold.
        Save all high quality isoforms to hq_isoforms_fa|fq if they are not None
        Save all low quality isoforms to lq_isoforms_fa|fq if they are not None
        """
        if guess_file_format(self.bas_fofn) != FILE_FORMATS.BAM:
            # Create input.fasta.fofn from bas_fofn
            self.add_log("Creating fasta fofn from bas/bax.h5/bam fofn",
                         level=logging.INFO)
            if self.fasta_fofn is None:
                self.fasta_fofn = op.join(self.nfl_dir, "input.fasta.fofn")
            self.add_log("fasta fofn={f}".format(f=self.fasta_fofn))
            convert_fofn_to_fasta(fofn_filename=self.bas_fofn,
                                  out_filename=self.fasta_fofn,
                                  fasta_out_dir=self.nfl_dir)
        else:
            self.fasta_fofn = None

        # Split non-full-length reads into smaller fasta files
        # and save files to root_dir/nfl_00.fasta, ..., .
        self.add_log("Splitting {nfl} into ".format(nfl=self.nfl_fa) +
                     "smaller files each containing {n} reads.".format(
                     n=self.ice_opts.nfl_reads_per_split),
                     level=logging.INFO)
        self._nfl_splitted_fas = splitFasta(input_fasta=self.nfl_fa,
                                            reads_per_split=self.ice_opts.nfl_reads_per_split,
                                            out_dir=self.nfl_dir,
                                            out_prefix="input.split")
        msg = "Splitted files are: " + "\n".join(self._nfl_splitted_fas)
        self.add_log(msg, level=logging.INFO)

        # Generating dazz DB for final.consensus.fasta
        ref_obj = DazzIDHandler(input_filename=self.final_consensus_fa, converted=False)
        ref_obj.make_db()
        msg = "Dazz DB made for: " + ref_obj.dazz_filename
        self.add_log(msg, level=logging.INFO)

        # Process nfl reads in each splitted fasta.
        self.add_log("Initializing IceAllPartials.", level=logging.INFO)

        self.icep = IceAllPartials(
            root_dir=self.root_dir,
            fasta_filenames=self._nfl_splitted_fas,
            ref_fasta=self.final_consensus_fa,
            out_pickle=self.nfl_all_pickle_fn,
            sge_opts=self.sge_opts,
            ccs_fofn=self.ccs_fofn)
        self.add_log("IceAllPartials log: {f}.".format(f=self.icep.log_fn),
                     level=logging.INFO)
        self.icep.run()
        self.add_log("IceAllPartials completed.", level=logging.INFO)

        self.add_log("Initializing IceQuiver.", level=logging.INFO)
        self.iceq = IceQuiver(root_dir=self.root_dir,
                              bas_fofn=self.bas_fofn,
                              fasta_fofn=self.fasta_fofn,
                              sge_opts=self.sge_opts,
                              tmp_dir=self.tmp_dir)
        self.add_log("IceQuiver log: {f}.".format(f=self.iceq.log_fn),
                     level=logging.INFO)
        self.iceq.run()
        self.add_log("IceQuiver finished.", level=logging.INFO)

        self.add_log("Initializing IceQuiverPostprocess.", level=logging.INFO)
        self.icepq = IceQuiverPostprocess(root_dir=self.root_dir,
                                          use_sge=self.sge_opts.use_sge,
                                          quit_if_not_done=False,
                                          ipq_opts=self.ipq_opts)
        self.add_log("IceQuiverPostprocess log: {f}.".
                     format(f=self.icepq.log_fn), level=logging.INFO)
        self.icepq.run()
        self.add_log("IceQuiverPostprocess finished.", level=logging.INFO)
Exemplo n.º 10
0
    def run(self):
        """Run"""
        iceq = IceQuiver(root_dir=self.root_dir, bas_fofn=None,
                         fasta_fofn=None, sge_opts=None,
                         prog_name="ice_quiver_merge")

        iceq.add_log(self.cmd_str())
        iceq.add_log("root_dir: {d}.".format(d=self.root_dir))
        iceq.add_log("Total number of chunks: N = {N}.".format(N=self.N))

        src = [iceq.submitted_quiver_jobs_log_of_chunk_i(i=i, num_chunks=self.N)
               for i in range(0, self.N)]
        for f in src:
            if not nfs_exists(f):
                raise IOError("Log {f} ".format(f=f) +
                              "of submitted quiver jobs does not exist.")

        dst = iceq.submitted_quiver_jobs_log

        iceq.add_log("Collecting submitted quiver jobs from:\n{src}\nto {dst}.".
                     format(src="\n".join(src), dst=dst))

        cat_files(src=src, dst=dst)

        iceq.close_log()
Exemplo n.º 11
0
 def cluster_dir(self, cid):
     """"overwrite IceQuiver.cluster_dir"""
     dir_name = IceQuiver.cluster_dir(self, cid)
     mkdir(dir_name)
     return dir_name
Exemplo n.º 12
0
    def run(self):
        """Run"""
        iceq = IceQuiver(root_dir=self.root_dir, bas_fofn=None,
                         fasta_fofn=None, sge_opts=None,
                         prog_name="ice_quiver_merge")

        iceq.add_log(self.cmd_str())
        iceq.add_log("root_dir: {d}.".format(d=self.root_dir))
        iceq.add_log("Total number of chunks: N = {N}.".format(N=self.N))

        src = [iceq.submitted_quiver_jobs_log_of_chunk_i(i=i, num_chunks=self.N)
               for i in range(0, self.N)]
        for f in src:
            if not nfs_exists(f):
                raise IOError("Log {f} ".format(f=f) +
                              "of submitted quiver jobs does not exist.")

        dst = iceq.submitted_quiver_jobs_log

        iceq.add_log("Collecting submitted quiver jobs from:\n{src}\nto {dst}.".
                     format(src="\n".join(src), dst=dst))

        cat_files(src=src, dst=dst)

        iceq.close_log()
Exemplo n.º 13
0
 def cluster_dir(self, cid):
     """"overwrite IceQuiver.cluster_dir"""
     dir_name = IceQuiver.cluster_dir(self, cid)
     mkdir(dir_name)
     return dir_name