Python CharonSession.seqrun_get方法代码示例

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def get_finished_seqruns_for_sample(project_id, sample_id,
                                    include_failed_libpreps=False):
    """Find all the finished seqruns for a particular sample.

    :param str project_id: The id of the project
    :param str sample_id: The id of the sample

    :returns: A dict of {libprep_01: [seqrun_01, ..., seqrun_nn], ...}
    :rtype: dict
    """
    charon_session = CharonSession()
    sample_libpreps = charon_session.sample_get_libpreps(projectid=project_id,
                                                         sampleid=sample_id)
    libpreps = collections.defaultdict(list)
    for libprep in sample_libpreps['libpreps']:
        if libprep.get('qc') != "FAILED" or include_failed_libpreps:
            libprep_id = libprep['libprepid']
            for seqrun in charon_session.libprep_get_seqruns(projectid=project_id,
                                                             sampleid=sample_id,
                                                             libprepid=libprep_id)['seqruns']:
                seqrun_id = seqrun['seqrunid']
                aln_status = charon_session.seqrun_get(projectid=project_id,
                                                       sampleid=sample_id,
                                                       libprepid=libprep_id,
                                                       seqrunid=seqrun_id).get('alignment_status')
                if aln_status == "DONE":
                    libpreps[libprep_id].append(seqrun_id)
                else:
                    LOG.debug('Skipping seqrun "{}" due to alignment_status '
                              '"{}"'.format(seqrun_id, aln_status))
        else:
            LOG.info('Skipping libprep "{}" due to qc status '
                     '"{}"'.format(libprep, libprep.get("qc")))
    return dict(libpreps)

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def check_for_preexisting_sample_runs(project_obj, sample_obj, restart_running_jobs, restart_finished_jobs):
    """If any analysis is undergoing or has completed for this sample's
    seqruns, raise a RuntimeError.

    :param NGIProject project_obj: The project object
    :param NGISample sample_obj: The sample object
    :param boolean restart_running_jobs: command line parameter
    :param boolean restart_finished_jobs: command line parameter

    :raise RuntimeError if the status is RUNNING or DONE and the flags do not allow to continue
    """
    project_id = project_obj.project_id
    sample_id = sample_obj.name
    charon_session = CharonSession()
    sample_libpreps = charon_session.sample_get_libpreps(projectid=project_id,
                                                         sampleid=sample_id)
    for libprep in sample_libpreps['libpreps']:
        libprep_id = libprep['libprepid']
        for seqrun in charon_session.libprep_get_seqruns(projectid=project_id,
                                                         sampleid=sample_id,
                                                         libprepid=libprep_id)['seqruns']:
            seqrun_id = seqrun['seqrunid']
            aln_status = charon_session.seqrun_get(projectid=project_id,
                                                   sampleid=sample_id,
                                                   libprepid=libprep_id,
                                                   seqrunid=seqrun_id).get('alignment_status')
            if (aln_status == "RUNNING" and not restart_running_jobs) or \
                (aln_status == "DONE" and not restart_finished_jobs):
                    raise RuntimeError('Project/Sample "{}/{}" has a preexisting '
                          'seqrun "{}" with status "{}"'.format(project_obj,
                          sample_obj, seqrun_id, aln_status))

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def write_to_charon_alignment_results(base_path, project_name, project_id, sample_id, libprep_id, seqrun_id):
    """Update the status of a sequencing run after alignment.

    :param str project_name: The name of the project (e.g. T.Durden_14_01)
    :param str project_id: The id of the project (e.g. P1171)
    :param str sample_id: ...
    :param str libprep_id: ...
    :param str seqrun_id: ...

    :raises RuntimeError: If the Charon database could not be updated
    :raises ValueError: If the output data could not be parsed.
    """
    charon_session = CharonSession()
    try:
        seqrun_dict = charon_session.seqrun_get(project_id, sample_id, libprep_id, seqrun_id)
    except CharonError as e:
        raise CharonError('Error accessing database for project "{}", sample {}; '
                           'could not update Charon while performing best practice: '
                           '{}'.format(project_name, sample_id,  e))
    piper_run_id = seqrun_id.split("_")[3]
    seqrun_dict["lanes"] = 0
    if seqrun_dict.get("alignment_status") == "DONE":
        LOG.warn("Sequencing run \"{}\" marked as DONE but writing new alignment results; "
                 "this will overwrite the previous results.".format(seqrun_id))
    # Find all the appropriate files
    piper_result_dir = os.path.join(base_path, "ANALYSIS", project_name, "02_preliminary_alignment_qc")
    try:
        os.path.isdir(piper_result_dir) and os.listdir(piper_result_dir)
    except OSError as e:
        raise ValueError("Piper result directory \"{}\" inaccessible when updating stats to Charon: {}.".format(piper_result_dir, e))
    piper_qc_dir_base = "{}.{}.{}".format(sample_id, piper_run_id, sample_id)
    piper_qc_path = "{}*/".format(os.path.join(piper_result_dir, piper_qc_dir_base))
    piper_qc_dirs = glob.glob(piper_qc_path)
    if not piper_qc_dirs: # Something went wrong in the alignment or we can't parse the file format
        raise ValueError("Piper qc directories under \"{}\" are missing or in an unexpected format when updating stats to Charon.".format(piper_qc_path))

    # Examine each lane and update the dict with its alignment metrics
    for qc_lane in piper_qc_dirs:
        genome_result = os.path.join(qc_lane, "genome_results.txt")
        # This means that if any of the lanes are missing results, the sequencing run is marked as a failure.
        # We should flag this somehow and send an email at some point.
        if not os.path.isfile(genome_result):
            raise ValueError("File \"genome_results.txt\" is missing from Piper result directory \"{}\"".format(piper_result_dir))
        # Get the alignment results for this lane
        lane_alignment_metrics = parse_qualimap_results(genome_result)
        # Update the dict for this lane
        update_seq_run_for_lane(seqrun_dict, lane_alignment_metrics)
    try:
        # Update the seqrun in the Charon database
        charon_session.seqrun_update(**seqrun_dict)
    except CharonError as e:
        error_msg = ('Failed to update run alignment status for run "{}" in project {} '
                     'sample {}, library prep {} to  Charon database: {}'.format(seqrun_id,
                      project_name, sample_id, libprep_id, e))
        raise CharonError(error_msg)

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def analyze(analysis_object, config=None, config_file_path=None):

    charon_session = CharonSession()
    charon_pj=charon_session.project_get(analysis_object.project.project_id)
    reference_genome=charon_pj.get('reference')
    if charon_pj.get("sequencing_facility") == "NGI-S":
        analysis_object.sequencing_facility="sthlm"
    elif charon_pj.get("sequencing_facility") == "NGI-U":
        analysis_object.sequencing_facility="upps"
    else:
        LOG.error("charon project not registered with stockholm or uppsala. Which config file should we use for the RNA pipeline ?")
        raise RuntimeError
    fastq_files=[]
    if reference_genome and reference_genome != 'other':
        for sample in analysis_object.project:
            try:
                charon_reported_status = charon_session.sample_get(analysis_object.project.project_id,
                                                                   sample).get('analysis_status')
                # Check Charon to ensure this hasn't already been processed
                do_analyze=handle_sample_status(analysis_object, sample, charon_reported_status)
                if not do_analyze :
                    continue
            except CharonError as e:
                LOG.error(e)

            for libprep in sample:
                charon_lp_status=charon_session.libprep_get(analysis_object.project.project_id, sample.name, libprep.name).get('qc')
                do_analyze=handle_libprep_status(analysis_object, libprep, charon_lp_status)
                if not do_analyze :
                    continue
                else:
                    for seqrun in libprep:
                        charon_sr_status=charon_session.seqrun_get(analysis_object.project.project_id, sample.name, libprep.name, seqrun.name).get('alignment_status')
                        do_analyze=handle_seqrun_status(analysis_object, seqrun, charon_sr_status)
                        if not do_analyze :
                            continue
                        else:
                            seqrun.being_analyzed=True
                            sample.being_analyzed = sample.being_analyzed or True
                            # filter out index files from analysis
                            for fastq_file in filter(lambda f: not is_index_file(f), seqrun.fastq_files):
                                fastq_path=os.path.join(analysis_object.project.base_path, "DATA", analysis_object.project.project_id, sample.name, libprep.name, seqrun.name, fastq_file)
                                fastq_files.append(fastq_path)
        
        if not fastq_files:
            LOG.error("No fastq files obtained for the analysis fo project {}, please check the Charon status.".format(analysis_object.project.name))
        else :
            if analysis_object.restart_running_jobs:
                stop_ongoing_analysis(analysis_object)
            fastq_dir=preprocess_analysis(analysis_object, fastq_files)
            sbatch_path=write_batch_job(analysis_object, reference_genome, fastq_dir)
            job_id=start_analysis(sbatch_path)
            analysis_path=os.path.join(analysis_object.project.base_path, "ANALYSIS", analysis_object.project.project_id, 'rna_ngi')
            record_project_job(analysis_object.project, job_id, analysis_path)

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def get_valid_seqruns_for_sample(project_id, sample_id,
                                 include_failed_libpreps=False,
                                 include_done_seqruns=False,
                                 status_field="alignment_status"):
    """Find all the valid seqruns for a particular sample.

    :param str project_id: The id of the project
    :param str sample_id: The id of the sample
    :param bool include_failed_libpreps: Include seqruns for libreps that have failed QC
    :param bool include_done_seqruns: Include seqruns that are already marked DONE

    :returns: A dict of {libprep_01: [seqrun_01, ..., seqrun_nn], ...}
    :rtype: dict

    :raises ValueError: If status_field is not a valid value
    """
    valid_status_values = ("alignment_status", "genotype_status",)
    if status_field not in valid_status_values:
        raise ValueError('"status_field" argument must be one of {} '
                         '(value passed was "{}")'.format(", ".join(valid_status_values),
                                                          status_field))
    charon_session = CharonSession()
    sample_libpreps = charon_session.sample_get_libpreps(projectid=project_id,
                                                         sampleid=sample_id)
    libpreps = collections.defaultdict(list)
    for libprep in sample_libpreps['libpreps']:
        if libprep.get('qc') != "FAILED" or include_failed_libpreps:
            libprep_id = libprep['libprepid']
            for seqrun in charon_session.libprep_get_seqruns(projectid=project_id,
                                                             sampleid=sample_id,
                                                             libprepid=libprep_id)['seqruns']:
                seqrun_id = seqrun['seqrunid']
                try:
                    aln_status = charon_session.seqrun_get(projectid=project_id,
                                                           sampleid=sample_id,
                                                           libprepid=libprep_id,
                                                           seqrunid=seqrun_id)[status_field]
                except KeyError:
                    LOG.error('Field "{}" not available for seqrun "{}" in Charon '
                              'for project "{}" / sample "{}". Including as '
                              'valid.'.format(status_field, seqrun_id,
                                              project_id, sample_id))
                    aln_status = None
                if aln_status != "DONE" or include_done_seqruns:
                    libpreps[libprep_id].append(seqrun_id)
                else:
                    LOG.info('Skipping seqrun "{}" due to {}'
                             '"{}"'.format(seqrun_id,status_field, aln_status))
        else:
            LOG.info('Skipping libprep "{}" due to qc status '
                     '"{}"'.format(libprep, libprep.get("qc")))
    return dict(libpreps)

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def update_charon_with_local_jobs_status():
    """Check the status of all locally-tracked jobs and update Charon accordingly.
    """
    LOG.info("Updating Charon with the status of all locally-tracked jobs...")
    with get_db_session() as session:
        charon_session = CharonSession()

        # Sequencing Run Analyses
        for seqrun_entry in session.query(SeqrunAnalysis).all():

            # Local names
            workflow = seqrun_entry.workflow
            project_name = seqrun_entry.project_name
            project_id = seqrun_entry.project_id
            project_base_path = seqrun_entry.project_base_path
            sample_id = seqrun_entry.sample_id
            libprep_id = seqrun_entry.libprep_id
            seqrun_id = seqrun_entry.seqrun_id
            pid = seqrun_entry.process_id

            exit_code = get_exit_code(workflow_name=workflow,
                                      project_base_path=project_base_path,
                                      project_name=project_name,
                                      sample_id=sample_id,
                                      libprep_id=libprep_id,
                                      seqrun_id=seqrun_id)
            label = "project/sample/libprep/seqrun {}/{}/{}/{}".format(project_name,
                                                                       sample_id,
                                                                       libprep_id,
                                                                       seqrun_id)
            try:
                if exit_code == 0:
                    # 0 -> Job finished successfully
                    LOG.info('Workflow "{}" for {} finished succesfully. '
                             'Recording status "DONE" in Charon'.format(workflow, label))
                    set_alignment_status = "DONE"
                    try:
                        write_to_charon_alignment_results(base_path=project_base_path,
                                                          project_name=project_name,
                                                          project_id=project_id,
                                                          sample_id=sample_id,
                                                          libprep_id=libprep_id,
                                                          seqrun_id=seqrun_id)
                    except (RuntimeError, ValueError) as e:
                        LOG.error(e)
                        set_alignment_status = "FAILED"
                    charon_session.seqrun_update(projectid=project_id,
                                                 sampleid=sample_id,
                                                 libprepid=libprep_id,
                                                 seqrunid=seqrun_id,
                                                 alignment_status=set_alignment_status)
                    # Job is only deleted if the Charon update succeeds
                    session.delete(seqrun_entry)
                elif exit_code == 1 or (not psutil.pid_exists(pid) and not exit_code):
                    if exit_code == 1:
                        # 1 -> Job failed (DATA_FAILURE / COMPUTATION_FAILURE ?)
                        LOG.info('Workflow "{}" for {} failed. Recording status '
                                 '"FAILED" in Charon.'.format(workflow, label))
                    else:
                        # Job failed without writing an exit code (process no longer running)
                        LOG.error('ERROR: No exit code found for process {} '
                                  'but it does not appear to be running '
                                  '(pid {} does not exist). Setting status to '
                                  '"FAILED", inspect manually'.format(label, pid))
                    charon_session.seqrun_update(projectid=project_id,
                                                 sampleid=sample_id,
                                                 libprepid=libprep_id,
                                                 seqrunid=seqrun_id,
                                                 alignment_status="FAILED")
                    # Job is only deleted if the Charon update succeeds
                    LOG.debug("Deleting local entry {}".format(seqrun_entry))
                    session.delete(seqrun_entry)
                else:
                    # None -> Job still running
                    charon_status = charon_session.seqrun_get(projectid=project_id,
                                                              sampleid=sample_id,
                                                              libprepid=libprep_id,
                                                              seqrunid=seqrun_id)['alignment_status']
                    if not charon_status == "RUNNING":
                        LOG.warn('Tracking inconsistency for {}: Charon status is "{}" but '
                                 'local process tracking database indicates it is running. '
                                 'Setting value in Charon to RUNNING.'.format(label, charon_status))
                        charon_session.seqrun_update(projectid=project_id,
                                                     sampleid=sample_id,
                                                     libprepid=libprep_id,
                                                     seqrunid=seqrun_id,
                                                     alignment_status="RUNNING")
            except CharonError as e:
                LOG.error('Unable to update Charon status for "{}": {}'.format(label, e))


        for sample_entry in session.query(SampleAnalysis).all():

            # Local names
            workflow = sample_entry.workflow
            project_name = sample_entry.project_name
            project_id = sample_entry.project_id
            project_base_path = sample_entry.project_base_path
            sample_id = sample_entry.sample_id
            pid = sample_entry.process_id
#.........这里部分代码省略.........

# 需要导入模块: from ngi_pipeline.database.classes import CharonSession [as 别名]
# 或者: from ngi_pipeline.database.classes.CharonSession import seqrun_get [as 别名]
def launch_analysis(level, projects_to_analyze, restart_failed_jobs=False,
                    config=None, config_file_path=None):
    """Launch the appropriate seqrun (flowcell-level) analysis for each fastq
    file in the project.

    :param list projects_to_analyze: The list of projects (Project objects) to analyze
    :param dict config: The parsed NGI configuration file; optional/has default.
    :param str config_file_path: The path to the NGI configuration file; optional/has default.
    """
    # Update Charon with the local state of all the jobs we're running
    update_charon_with_local_jobs_status()
    charon_session = CharonSession()
    for project in projects_to_analyze:
        # Get information from Charon regarding which workflows to run
        try:
            # E.g. "NGI" for NGI DNA Samples
            workflow = charon_session.project_get(project.project_id)["pipeline"]
        except (KeyError, CharonError) as e:
            # Workflow missing from Charon?
            LOG.error('Skipping project "{}" because of error: {}'.format(project, e))
            continue
        try:
            analysis_engine_module_name = config["analysis"]["workflows"][workflow]["analysis_engine"]
        except KeyError:
            error_msg = ("No analysis engine for workflow \"{}\" specified "
                         "in configuration file. Skipping this workflow "
                         "for project {}".format(workflow, project))
            LOG.error(error_msg)
            raise RuntimeError(error_msg)
        # Import the adapter module specified in the config file (e.g. piper_ngi)
        try:
            analysis_module = importlib.import_module(analysis_engine_module_name)
        except ImportError as e:
            error_msg = ('Skipping project "{}" workflow "{}": couldn\'t import '
                         'module "{}": {}'.format(project, workflow, analysis_engine_module_name, e))
            LOG.error(error_msg)
            # Next project
            continue

        # This is weird
        objects_to_process = []
        if level == "sample":
            for sample in project:
                objects_to_process.append({"project": project, "sample": sample})
        elif level == "seqrun":
            for sample in project:
                for libprep in sample:
                    for seqrun in libprep:
                        objects_to_process.append({"project": project,
                                                   "sample": sample,
                                                   "libprep": libprep,
                                                   "seqrun": seqrun})
        # Still weird and not so great
        for obj_dict in objects_to_process:
            project = obj_dict.get("project")
            sample = obj_dict.get("sample")
            libprep = obj_dict.get("libprep")
            seqrun = obj_dict.get("seqrun")

            try:
                if level == "seqrun":
                    charon_reported_status = charon_session.seqrun_get(project.project_id,
                                                                       sample, libprep,
                                                                       seqrun)['alignment_status']
                else: # sample-level
                    charon_reported_status = charon_session.sample_get(project.project_id,
                                                                       sample)['status']
            except (CharonError, KeyError) as e:
                LOG.warn('Unable to get required information from Charon for '
                          'sample "{}" / project "{}" -- forcing it to new: {}'.format(sample, project, e))
                if level == "seqrun":
                    charon_session.seqrun_update(project.project_id, sample.name, libprep.name, seqrun.name, alignment_status="NEW")
                    charon_reported_status = charon_session.seqrun_get(project.project_id,
                                                                       sample, libprep,
                                                                       seqrun)['alignment_status']
                else:
                    charon_session.sample_update(project.project_id, sample.name, status="NEW")
                    charon_reported_status = charon_session.sample_get(project.project_id,
                                                                       sample)['status']

            # Check Charon to ensure this hasn't already been processed
            if charon_reported_status in ("RUNNING", "DONE"):
                if level == "seqrun":
                    LOG.info('Charon reports seqrun analysis for project "{}" / sample "{}" '
                             '/ libprep "{}" / seqrun "{}" does not need processing '
                             ' (already "{}")'.format(project, sample, libprep, seqrun,
                                                      charon_reported_status))
                else: # Sample
                    LOG.info('Charon reports seqrun analysis for project "{}" / sample "{}" '
                             'does not need processing '
                             ' (already "{}")'.format(project, sample, charon_reported_status))
                continue
            elif charon_reported_status == "FAILED":
                if not restart_failed_jobs:
                    if level == "seqrun":
                        LOG.error('FAILED:  Project "{}" / sample "{}" / library "{}" '
                                  '/ flowcell "{}": Charon reports FAILURE, manual '
                                  'investigation needed!'.format(project, sample, libprep, seqrun))
                    else: # Sample
                        LOG.error('FAILED:  Project "{}" / sample "{}" Charon reports FAILURE, manual '
#.........这里部分代码省略.........