예제 #1
0
def mutect_caller(align_bams, items, ref_file, assoc_files, region=None,
                  out_file=None):
    """Run the MuTect paired analysis algorithm.
    """
    config = items[0]["config"]
    if out_file is None:
        out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
    if not file_exists(out_file):
        base_config = items[0]["config"]
        broad_runner = broad.runner_from_config(base_config, "mutect")
        out_file_mutect = (out_file.replace(".vcf", "-mutect.vcf")
                           if "vcf" in out_file else out_file + "-mutect.vcf")
        broad_runner, params = \
            _mutect_call_prep(align_bams, items, ref_file, assoc_files,
                                   region, out_file_mutect)
        if (not isinstance(region, (list, tuple)) and
              not all(has_aligned_reads(x, region) for x in align_bams)):
                vcfutils.write_empty_vcf(out_file)
                return
        with file_transaction(out_file_mutect) as tx_out_file:
            # Rationale: MuTect writes another table to stdout, which we don't need
            params += ["--vcf", tx_out_file, "-o", os.devnull]
            broad_runner.run_mutect(params)
        _rename_allelic_fraction_field(out_file_mutect,config)
        disable_SID = True # SID isn't great, so use Scalpel instead
        if "appistry" not in broad_runner.get_mutect_version() or disable_SID:
            # Scalpel InDels
            is_paired = "-I:normal" in params
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file + "-somaticIndels.vcf")
            if scalpel.is_installed(items[0]["config"]):
                with file_transaction(out_file_indels) as tx_out_file2:
                    if not is_paired:
                        scalpel._run_scalpel_caller(align_bams, items, ref_file, assoc_files,
                                                    region=region, out_file=tx_out_file2)
                    else:
                        scalpel._run_scalpel_paired(align_bams, items, ref_file, assoc_files,
                                                    region=region, out_file=tx_out_file2)
                out_file = vcfutils.combine_variant_files(orig_files=[out_file_mutect, out_file_indels],
                                                          out_file=out_file,
                                                          ref_file=items[0]["sam_ref"],
                                                          config=items[0]["config"],
                                                          region=region)
            else:
                utils.symlink_plus(out_file_mutect, out_file)
        else:
            # SomaticIndelDetector modifications
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file + "-somaticIndels.vcf")
            params_indels = _SID_call_prep(align_bams, items, ref_file, assoc_files,
                                           region, out_file_indels)
            with file_transaction(out_file_indels) as tx_out_file:
                params_indels += ["-o", tx_out_file]
                broad_runner.run_mutect(params_indels)
            out_file = vcfutils.combine_variant_files(orig_files=[out_file_mutect, out_file_indels],
                                                      out_file=out_file,
                                                      ref_file=items[0]["sam_ref"],
                                                      config=items[0]["config"],
                                                      region=region)
    return out_file
예제 #2
0
def mutect_caller(align_bams, items, ref_file, assoc_files, region=None,
                  out_file=None):
    """Run the MuTect paired analysis algorithm.
    """
    config = items[0]["config"]
    if out_file is None:
        out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
    if not file_exists(out_file):
        base_config = items[0]["config"]
        broad_runner = broad.runner_from_config(base_config, "mutect")
        out_file_mutect = (out_file.replace(".vcf", "-mutect.vcf")
                           if "vcf" in out_file else out_file + "-mutect.vcf")
        broad_runner, params = \
            _mutect_call_prep(align_bams, items, ref_file, assoc_files,
                                   region, out_file_mutect)
        if (not isinstance(region, (list, tuple)) and
              not all(has_aligned_reads(x, region) for x in align_bams)):
                vcfutils.write_empty_vcf(out_file)
                return
        out_file_orig = "%s-orig%s" % utils.splitext_plus(out_file_mutect)
        with file_transaction(config, out_file_orig) as tx_out_file:
            # Rationale: MuTect writes another table to stdout, which we don't need
            params += ["--vcf", tx_out_file, "-o", os.devnull]
            broad_runner.run_mutect(params)
        is_paired = "-I:normal" in params
        out_file_mutect = _fix_mutect_output(out_file_orig, config, out_file_mutect, is_paired)
        indelcaller = vcfutils.get_indelcaller(base_config)
        if "scalpel" in indelcaller.lower():
            # Scalpel InDels
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file + "-somaticIndels.vcf")
            if scalpel.is_installed(items[0]["config"]):
                with file_transaction(config, out_file_indels) as tx_out_file2:
                    if not is_paired:
                        vcfutils.check_paired_problems(items)
                        scalpel._run_scalpel_caller(align_bams, items, ref_file, assoc_files,
                                                    region=region, out_file=tx_out_file2)
                    else:
                        scalpel._run_scalpel_paired(align_bams, items, ref_file, assoc_files,
                                                    region=region, out_file=tx_out_file2)
                out_file = vcfutils.combine_variant_files(orig_files=[out_file_mutect, out_file_indels],
                                                          out_file=out_file,
                                                          ref_file=items[0]["sam_ref"],
                                                          config=items[0]["config"],
                                                          region=region)
            else:
                utils.symlink_plus(out_file_mutect, out_file)
        elif "pindel" in indelcaller.lower():
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file + "-somaticIndels.vcf")
            if pindel.is_installed(items[0]["config"]):
                pindel._run_tumor_pindel_caller(align_bams, items, ref_file, assoc_files, region=region,
                                          out_file=out_file_indels)
                out_file = vcfutils.combine_variant_files(orig_files=[out_file_mutect, out_file_indels],
                                                          out_file=out_file,
                                                          ref_file=ref_file,
                                                          config=items[0]["config"],
                                                          region=region)
            else:
                utils.symlink_plus(out_file_mutect, out_file)
        elif (("somaticindeldetector" in indelcaller.lower() or "sid" in indelcaller.lower())
              and "appistry" in broad_runner.get_mutect_version()):
            # SomaticIndelDetector InDels
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file + "-somaticIndels.vcf")
            params_indels = _SID_call_prep(align_bams, items, ref_file, assoc_files,
                                           region, out_file_indels)
            with file_transaction(config, out_file_indels) as tx_out_file:
                params_indels += ["-o", tx_out_file]
                broad_runner.run_mutect(params_indels)
            out_file = vcfutils.combine_variant_files(orig_files=[out_file_mutect, out_file_indels],
                                                      out_file=out_file,
                                                      ref_file=items[0]["sam_ref"],
                                                      config=items[0]["config"],
                                                      region=region)
        else:
            utils.symlink_plus(out_file_mutect, out_file)
    return out_file
예제 #3
0
def mutect_caller(align_bams,
                  items,
                  ref_file,
                  assoc_files,
                  region=None,
                  out_file=None):
    """Run the MuTect paired analysis algorithm.
    """
    config = items[0]["config"]
    if out_file is None:
        out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(
            align_bams[0])[0]
    if not file_exists(out_file):
        base_config = items[0]["config"]
        broad_runner = broad.runner_from_config(base_config, "mutect")
        out_file_mutect = (out_file.replace(".vcf", "-mutect.vcf")
                           if "vcf" in out_file else out_file + "-mutect.vcf")
        broad_runner, params = \
            _mutect_call_prep(align_bams, items, ref_file, assoc_files,
                                   region, out_file_mutect)
        if (not isinstance(region, (list, tuple))
                and not all(has_aligned_reads(x, region) for x in align_bams)):
            vcfutils.write_empty_vcf(out_file)
            return
        out_file_orig = "%s-orig%s" % utils.splitext_plus(out_file_mutect)
        if not file_exists(out_file_orig):
            with file_transaction(config, out_file_orig) as tx_out_file:
                # Rationale: MuTect writes another table to stdout, which we don't need
                params += ["--vcf", tx_out_file, "-o", os.devnull]
                broad_runner.run_mutect(params)
        is_paired = "-I:normal" in params
        if not utils.file_uptodate(out_file_mutect, out_file_orig):
            out_file_mutect = _fix_mutect_output(out_file_orig, config,
                                                 out_file_mutect, is_paired)
        indelcaller = vcfutils.get_indelcaller(base_config)
        if ("scalpel" in indelcaller.lower() and region
                and isinstance(region, (tuple, list))
                and chromhacks.is_autosomal_or_sex(region[0])):
            # Scalpel InDels
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file +
                               "-somaticIndels.vcf")
            if scalpel.is_installed(items[0]["config"]):
                if not is_paired:
                    vcfutils.check_paired_problems(items)
                    scalpel._run_scalpel_caller(align_bams,
                                                items,
                                                ref_file,
                                                assoc_files,
                                                region=region,
                                                out_file=out_file_indels)
                else:
                    scalpel._run_scalpel_paired(align_bams,
                                                items,
                                                ref_file,
                                                assoc_files,
                                                region=region,
                                                out_file=out_file_indels)
                out_file = vcfutils.combine_variant_files(
                    orig_files=[out_file_mutect, out_file_indels],
                    out_file=out_file,
                    ref_file=items[0]["sam_ref"],
                    config=items[0]["config"],
                    region=region)
            else:
                utils.symlink_plus(out_file_mutect, out_file)
        elif "pindel" in indelcaller.lower():
            from bcbio.structural import pindel
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file +
                               "-somaticIndels.vcf")
            if pindel.is_installed(items[0]["config"]):
                pindel._run_tumor_pindel_caller(align_bams,
                                                items,
                                                ref_file,
                                                assoc_files,
                                                region=region,
                                                out_file=out_file_indels)
                out_file = vcfutils.combine_variant_files(
                    orig_files=[out_file_mutect, out_file_indels],
                    out_file=out_file,
                    ref_file=ref_file,
                    config=items[0]["config"],
                    region=region)
            else:
                utils.symlink_plus(out_file_mutect, out_file)
        elif (("somaticindeldetector" in indelcaller.lower()
               or "sid" in indelcaller.lower())
              and "appistry" in broad_runner.get_mutect_version()):
            # SomaticIndelDetector InDels
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file +
                               "-somaticIndels.vcf")
            params_indels = _SID_call_prep(align_bams, items, ref_file,
                                           assoc_files, region,
                                           out_file_indels)
            with file_transaction(config, out_file_indels) as tx_out_file:
                params_indels += ["-o", tx_out_file]
                broad_runner.run_mutect(params_indels)
            out_file = vcfutils.combine_variant_files(
                orig_files=[out_file_mutect, out_file_indels],
                out_file=out_file,
                ref_file=items[0]["sam_ref"],
                config=items[0]["config"],
                region=region)
        else:
            utils.symlink_plus(out_file_mutect, out_file)
    return out_file
예제 #4
0
def mutect_caller(align_bams,
                  items,
                  ref_file,
                  assoc_files,
                  region=None,
                  out_file=None):
    """Run the MuTect paired analysis algorithm.
    """
    if out_file is None:
        out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(
            align_bams[0])[0]
    if not file_exists(out_file):
        base_config = items[0]["config"]
        broad_runner = broad.runner_from_config(base_config, "mutect")
        out_file_mutect = (out_file.replace(".vcf", "-mutect.vcf")
                           if "vcf" in out_file else out_file + "-mutect.vcf")
        broad_runner, params = \
            _mutect_call_prep(align_bams, items, ref_file, assoc_files,
                                   region, out_file_mutect)
        if (not isinstance(region, (list, tuple))
                and not all(has_aligned_reads(x, region) for x in align_bams)):
            vcfutils.write_empty_vcf(out_file)
            return
        with file_transaction(out_file_mutect) as tx_out_file:
            # Rationale: MuTect writes another table to stdout, which we don't need
            params += ["--vcf", tx_out_file, "-o", os.devnull]
            broad_runner.run_mutect(params)
        if "appistry" not in broad_runner.get_mutect_version():
            # Scalpel InDels
            is_paired = "-I:normal" in params
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file +
                               "-somaticIndels.vcf")
            if scalpel.is_installed(items[0]["config"]):
                with file_transaction(out_file_indels) as tx_out_file2:
                    if not is_paired:
                        scalpel._run_scalpel_caller(align_bams,
                                                    items,
                                                    ref_file,
                                                    assoc_files,
                                                    region=region,
                                                    out_file=tx_out_file2)
                    else:
                        scalpel._run_scalpel_paired(align_bams,
                                                    items,
                                                    ref_file,
                                                    assoc_files,
                                                    region=region,
                                                    out_file=tx_out_file2)
                out_file = vcfutils.combine_variant_files(
                    orig_files=[out_file_mutect, out_file_indels],
                    out_file=out_file,
                    ref_file=items[0]["sam_ref"],
                    config=items[0]["config"],
                    region=region)
            else:
                utils.symlink_plus(out_file_mutect, out_file)
        else:
            # SomaticIndelDetector modifications
            out_file_indels = (out_file.replace(".vcf", "-somaticIndels.vcf")
                               if "vcf" in out_file else out_file +
                               "-somaticIndels.vcf")
            params_indels = _SID_call_prep(align_bams, items, ref_file,
                                           assoc_files, region,
                                           out_file_indels)
            with file_transaction(out_file_indels) as tx_out_file:
                params_indels += ["-o", tx_out_file]
                broad_runner.run_mutect(params_indels)
            out_file = vcfutils.combine_variant_files(
                orig_files=[out_file_mutect, out_file_indels],
                out_file=out_file,
                ref_file=items[0]["sam_ref"],
                config=items[0]["config"],
                region=region)
    return out_file