def salmon_index(gtf_file, ref_file, data, out_dir):
out_dir = os.path.join(out_dir, "index", sailfish.get_build_string(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambiguate(data))
salmon = config_utils.get_program("salmon", dd.get_config(data))
num_cores = dd.get_num_cores(data)
if dd.get_transcriptome_fasta(data):
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data)
assert file_exists(gtf_fa), "%s was not found, exiting." % gtf_fa
tmpdir = dd.get_tmp_dir(data)
out_file = os.path.join(out_dir, "versionInfo.json")
if file_exists(out_file):
logger.info("Transcriptome index for %s detected, skipping building." %
gtf_fa)
return out_dir
files = dd.get_input_sequence_files(data)
readlength = bam.fastq.estimate_read_length(files[0])
if readlength % 2 == 0:
readlength -= 1
kmersize = min(readlength, 31)
with file_transaction(data, out_dir) as tx_out_dir:
cmd = "{salmon} index -k {kmersize} -p {num_cores} -i {tx_out_dir} -t {gtf_fa}"
message = "Creating Salmon index for {gtf_fa}."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
if dd.get_transcriptome_align(data) and not dd.get_transcriptome_bam(data):
file1, file2 = None, None
if dd.get_disambiguate(data):
bam_path = data["work_bam"]
fastq_paths = alignprep._bgzip_from_bam(bam_path, data["dirs"], data["config"], is_retry=False, output_infix='-transcriptome')
if len(fastq_paths) == 2:
file1, file2 = fastq_paths
else:
file1, file2 = fastq_paths[0], None
else:
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("Transcriptome alignment was flagged to run, but the "
"transcriptome BAM file was not found. Aligning to the "
"transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def convert_to_kallisto(data):
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename, "fastq")
out_file = os.path.join(kallisto_dir, "barcodes.batch")
umis = config_utils.get_program("umis", dd.get_config(data))
if file_exists(out_file):
return out_file
if dd.get_minimum_barcode_depth(data):
cb_histogram = os.path.join(work_dir, "umis", samplename,
"cb-histogram.txt")
cb_cutoff = dd.get_minimum_barcode_depth(data)
cb_options = "--cb_histogram {cb_histogram} --cb_cutoff {cb_cutoff}"
cb_options = cb_options.format(**locals())
else:
cb_options = ""
cmd = ("{umis} kallisto {cb_options} --out_dir {tx_kallisto_dir} {fq1}")
with file_transaction(data, kallisto_dir) as tx_kallisto_dir:
safe_makedir(tx_kallisto_dir)
message = ("Transforming %s to Kallisto singlecell format. " % fq1)
do.run(cmd.format(**locals()), message)
return out_file
def convert_to_kallisto(data):
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename, "fastq")
out_file = os.path.join(kallisto_dir, "barcodes.batch")
umis = config_utils.get_program("umis", dd.get_config(data))
if file_exists(out_file):
return out_file
if dd.get_minimum_barcode_depth(data):
cb_histogram = os.path.join(work_dir, "umis", samplename, "cb-histogram.txt")
cb_cutoff = dd.get_minimum_barcode_depth(data)
cb_options = "--cb_histogram {cb_histogram} --cb_cutoff {cb_cutoff}"
cb_options = cb_options.format(**locals())
else:
cb_options = ""
cmd = ("{umis} kallisto {cb_options} --out_dir {tx_kallisto_dir} {fq1}")
with file_transaction(data, kallisto_dir) as tx_kallisto_dir:
safe_makedir(tx_kallisto_dir)
message = ("Transforming %s to Kallisto singlecell format. "
% fq1)
do.run(cmd.format(**locals()), message)
return out_file
def salmon_decoy_index(gtf_file, data, out_dir):
input_dir = os.path.join(dd.get_work_dir(data), "inputs", "transcriptome")
decoy_transcriptome = os.path.join(
input_dir,
sailfish.get_build_string(data) + "-decoy.fa")
out_dir = os.path.join(out_dir, "index", sailfish.get_build_string(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambiguate(data))
salmon = config_utils.get_program("salmon", dd.get_config(data))
num_cores = dd.get_num_cores(data)
if dd.get_transcriptome_fasta(data):
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data)
assert file_exists(gtf_fa), "%s was not found, exiting." % gtf_fa
decoy_sequence_file = get_decoy_sequence_file(data)
decoy_name_file = get_decoy_name_file(data)
gtf_fa = create_decoy_transcriptome(gtf_fa, get_decoy_sequence_file(data),
decoy_transcriptome)
out_file = os.path.join(out_dir, "versionInfo.json")
if file_exists(out_file):
logger.info("Transcriptome index for %s detected, skipping building." %
gtf_fa)
return out_dir
files = dd.get_input_sequence_files(data)
kmersize = sailfish.pick_kmersize(files[0])
with file_transaction(data, out_dir) as tx_out_dir:
cmd = (
"{salmon} index -k {kmersize} -p {num_cores} -i {tx_out_dir} -t {gtf_fa} "
"--decoys {decoy_name_file} ")
message = "Creating decoy-aware Salmon index for {gtf_fa} with {kmersize} bp kmers."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def rapmap_index(gtf_file, ref_file, algorithm, data, out_dir):
valid_indexes = ["pseudoindex", "quasiindex"]
index_type = algorithm + "index"
assert index_type in valid_indexes, \
"RapMap only supports %s indices." % valid_indexes
out_dir = os.path.join(out_dir, index_type, dd.get_genome_build(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambguate(data))
rapmap = config_utils.get_program("rapmap", dd.get_config(data))
# use user supplied transcriptome FASTA file if it exists
if dd.get_transcriptome_fasta(data):
out_dir = os.path.join(out_dir, index_type, dd.get_genome_build(data))
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data)
tmpdir = dd.get_tmp_dir(data)
if file_exists(out_dir + "rapidx.jfhash"):
return out_dir
files = dd.get_input_sequence_files(data)
kmersize = sailfish.pick_kmersize(files[0])
message = "Creating rapmap {index_type} for {gtf_fa} with {kmersize} bp kmers."
with file_transaction(out_dir) as tx_out_dir:
cmd = "{rapmap} {index_type} -k {kmersize} -i {tx_out_dir} -t {gtf_fa}"
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_transcriptome_align(data):
# to create a disambiguated transcriptome file realign with bowtie2
if dd.get_disambiguate(data):
logger.info("Aligning to the transcriptome with bowtie2 using the "
"disambiguated reads.")
bam_path = data["work_bam"]
fastq_paths = alignprep._bgzip_from_bam(
bam_path,
data["dirs"],
data,
is_retry=False,
output_infix='-transcriptome')
if len(fastq_paths) == 2:
file1, file2 = fastq_paths
else:
file1, file2 = fastq_paths[0], None
ref_file = dd.get_ref_file(data)
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
else:
file1, file2 = dd.get_input_sequence_files(data)
if not dd.get_transcriptome_bam(data):
ref_file = dd.get_ref_file(data)
logger.info(
"Transcriptome alignment was flagged to run, but the "
"transcriptome BAM file was not found. Aligning to the "
"transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
data = spikein.counts_spikein(data)
return [[data]]
def umi_transform(data):
"""
transform each read by identifying the barcode and UMI for each read
and putting the information in the read name
"""
fq1, fq2 = dd.get_input_sequence_files(data)
fq2 = fq2 if fq2 else ""
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
transform = dd.get_umi_type(data)
transform_data = transforms[transform]
safe_makedir(umi_dir)
transform_file = os.path.join(umi_dir, transform + ".json")
transform_file = write_transform_file(transform_data, transform_file)
out_base = dd.get_sample_name(data) + ".umitransformed.fq.gz"
out_file = os.path.join(umi_dir, out_base)
if file_exists(out_file):
data["files"] = [out_file]
return [[data]]
index_option = "--dual_index" if transform_data["dual"] else ""
if len(dd.get_cellular_barcodes(data)) == 2:
split_option = "--separate_cb"
else:
split_option = ""
umis = config_utils.get_program("umis", data, default="umis")
cmd = (
"{umis} fastqtransform {index_option} {split_option} {transform_file} "
"{fq1} {fq2} "
"| seqtk seq -L 20 - | gzip > {tx_out_file}")
message = (
"Inserting UMI and barcode information into the read name of %s" % fq1)
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()), message)
data["files"] = [out_file]
return [[data]]
def salmon_index(gtf_file, ref_file, data, out_dir):
out_dir = os.path.join(out_dir, "index", sailfish.get_build_string(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambiguate(data))
salmon = config_utils.get_program("salmon", dd.get_config(data))
num_cores = dd.get_num_cores(data)
if dd.get_transcriptome_fasta(data):
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data, out_dir)
assert file_exists(gtf_fa), "%s was not found, exiting." % gtf_fa
tmpdir = dd.get_tmp_dir(data)
out_file = os.path.join(out_dir, "versionInfo.json")
if file_exists(out_file):
return out_dir
files = dd.get_input_sequence_files(data)
readlength = bam.fastq.estimate_read_length(files[0])
if readlength % 2 == 0:
readlength -= 1
kmersize = min(readlength, 31)
with file_transaction(data, out_dir) as tx_out_dir:
cmd = "{salmon} index -k {kmersize} -p {num_cores} -i {tx_out_dir} -t {gtf_fa}"
message = "Creating Salmon index for {gtf_fa}."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def counts_spikein(data):
data = utils.to_single_data(data)
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "spikein", samplename)
fasta_file = dd.get_spikein_fasta(data)
if not fasta_file:
return data
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
readlength = fastq.estimate_read_length(fq1)
if readlength % 2 == 0:
readlength -= 1
kmersize = min(readlength, 31)
logger.info("kmersize used for salmon index at spikein quant: %s" %
kmersize)
kmersize = kmersize if not dd.get_analysis(
data).lower() == "smallrna-seq" else 15
fasta_index = _index_spikein(fasta_file, salmon_dir, data, kmersize)
out_file = _salmon_quant_reads(fq1, fq2, salmon_dir, fasta_index, data)
data = dd.set_spikein_counts(data, out_file)
return data
def umi_transform(data):
"""
transform each read by identifying the barcode and UMI for each read
and putting the information in the read name
"""
fq1, fq2 = dd.get_input_sequence_files(data)
fq2 = fq2 if fq2 else ""
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
transform = dd.get_umi_type(data)
transform_data = transforms[transform]
safe_makedir(umi_dir)
transform_file = os.path.join(umi_dir, transform + ".json")
transform_file = write_transform_file(transform_data, transform_file)
out_base = dd.get_sample_name(data) + ".umitransformed.fq.gz"
out_file = os.path.join(umi_dir, out_base)
if file_exists(out_file):
data["files"] = [out_file]
return [[data]]
index_option = "--dual_index" if transform_data["dual"] else ""
if len(dd.get_cellular_barcodes(data)) == 2:
split_option = "--separate_cb"
else:
split_option = ""
umis = config_utils.get_program("umis", data, default="umis")
cmd = ("{umis} fastqtransform {index_option} {split_option} {transform_file} "
"{fq1} {fq2} "
"| seqtk seq -L 20 - | gzip > {tx_out_file}")
message = ("Inserting UMI and barcode information into the read name of %s"
% fq1)
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()), message)
data["files"] = [out_file]
return [[data]]
def trim_adapters(data):
fq1, fq2 = dd.get_input_sequence_files(data)
skewer = config_utils.get_program("skewer", data, default="skewer")
nthreads = dd.get_num_cores(data)
samplename = dd.get_sample_name(data)
out_dir = safe_makedir(
os.path.join(dd.get_work_dir(data), "trimmed", samplename))
of1 = os.path.join(out_dir, samplename + "-trimmed-pair1.fastq.gz")
of2 = os.path.join(out_dir, samplename + "-trimmed-pair2.fastq.gz")
of2 = of2 if fq2 else None
if fq1 and fq2:
if file_exists(of1) and file_exists(of2):
return of1, of2
else:
if file_exists(of1):
return of1, None
safe_makedir(out_dir)
file_string = "{fq1} {fq2} " if fq2 else "{fq1} "
fw_cmd = _fw_command(data)
rv_cmd = _rv_command(data)
mode = "tail" if not fq2 else "pe"
cmd = ("{skewer} --min 25 --threads {nthreads} -q 5 "
"{fw_cmd} "
"{rv_cmd} "
"-m {mode} "
"--compress --output {out_stem} ") + file_string
with file_transaction(out_dir) as tx_out_dir:
safe_makedir(tx_out_dir)
out_stem = os.path.join(tx_out_dir, samplename)
message = "Trimming {fq1}, {fq2} with skewer.".format(**locals())
do.run(cmd.format(**locals()), message)
return of1, of2
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
if dd.get_transcriptome_align(data) and not dd.get_transcriptome_bam(data):
file1, file2 = None, None
if dd.get_disambiguate(data):
bam_path = data["work_bam"]
fastq_paths = alignprep._bgzip_from_bam(
bam_path,
data["dirs"],
data["config"],
is_retry=False,
output_infix='-transcriptome')
if len(fastq_paths) == 2:
file1, file2 = fastq_paths
else:
file1, file2 = fastq_paths[0], None
else:
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("Transcriptome alignment was flagged to run, but the "
"transcriptome BAM file was not found. Aligning to the "
"transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def trim_adapters(data):
fq1, fq2 = dd.get_input_sequence_files(data)
skewer = config_utils.get_program("skewer", data, default="skewer")
nthreads = dd.get_num_cores(data)
samplename = dd.get_sample_name(data)
out_dir = os.path.join(dd.get_work_dir(data), "trimmed", samplename)
of1 = os.path.join(out_dir, samplename + "-trimmed-pair1.fastq.gz")
of2 = os.path.join(out_dir, samplename + "-trimmed-pair2.fastq.gz")
of2 = of2 if fq2 else None
if fq1 and fq2:
if file_exists(of1) and file_exists(of2):
return of1, of2
else:
if file_exists(of1):
return of1, None
safe_makedir(out_dir)
file_string = "{fq1} {fq2} " if fq2 else "{fq1} "
fw_cmd = _fw_command(data)
rv_cmd = _rv_command(data)
mode = "tail" if not fq2 else "pe"
cmd = ("{skewer} --min 25 --threads {nthreads} -q 5 "
"{fw_cmd} "
"{rv_cmd} "
"-m {mode} "
"--compress --output {out_stem} ") + file_string
with file_transaction(out_dir) as tx_out_dir:
safe_makedir(tx_out_dir)
out_stem = os.path.join(tx_out_dir, samplename)
message = "Trimming {fq1}, {fq2} with skewer.".format(**locals())
do.run(cmd.format(**locals()), message)
return of1, of2
def rapmap_index(gtf_file, ref_file, algorithm, data, out_dir):
valid_indexes = ["pseudoindex", "quasiindex"]
index_type = algorithm + "index"
assert index_type in valid_indexes, \
"RapMap only supports %s indices." % valid_indexes
out_dir = os.path.join(out_dir, index_type, dd.get_genome_build(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambguate(data))
rapmap = config_utils.get_program("rapmap", dd.get_config(data))
# use user supplied transcriptome FASTA file if it exists
if dd.get_transcriptome_fasta(data):
out_dir = os.path.join(out_dir, index_type, dd.get_genome_build(data))
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data)
tmpdir = dd.get_tmp_dir(data)
if file_exists(out_dir + "rapidx.jfhash"):
return out_dir
files = dd.get_input_sequence_files(data)
kmersize = sailfish.pick_kmersize(files[0])
message = "Creating rapmap {index_type} for {gtf_fa} with {kmersize} bp kmers."
with file_transaction(out_dir) as tx_out_dir:
cmd = "{rapmap} {index_type} -k {kmersize} -i {tx_out_dir} -t {gtf_fa}"
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def process_alignment(data, alt_input=None):
"""Do an alignment of fastq files, preparing a sorted BAM output file.
"""
fastq1, fastq2 = dd.get_input_sequence_files(data)
if alt_input:
fastq1, fastq2 = alt_input
config = data["config"]
aligner = config["algorithm"].get("aligner", None)
if fastq1 and objectstore.file_exists_or_remote(fastq1) and aligner:
logger.info("Aligning lane %s with %s aligner" %
(data["rgnames"]["lane"], aligner))
data = align_to_sort_bam(fastq1, fastq2, aligner, data)
data = _add_supplemental_bams(data)
elif fastq1 and os.path.exists(fastq1) and fastq1.endswith(".bam"):
sort_method = config["algorithm"].get("bam_sort")
bamclean = config["algorithm"].get("bam_clean")
if bamclean is True or bamclean == "picard":
if sort_method and sort_method != "coordinate":
raise ValueError(
"Cannot specify `bam_clean: picard` with `bam_sort` other than coordinate: %s"
% sort_method)
out_bam = cleanbam.picard_prep(fastq1, data["rgnames"],
data["sam_ref"], data["dirs"], data)
elif sort_method:
runner = broad.runner_from_config(config)
out_file = os.path.join(
data["dirs"]["work"], "{}-sort.bam".format(
os.path.splitext(os.path.basename(fastq1))[0]))
out_bam = runner.run_fn("picard_sort", fastq1, sort_method,
out_file)
else:
out_bam = link_bam_file(
fastq1,
os.path.join(data["dirs"]["work"], "prealign",
data["rgnames"]["sample"]))
bam.index(out_bam, data["config"])
bam.check_header(out_bam, data["rgnames"], data["sam_ref"],
data["config"])
dedup_bam = postalign.dedup_bam(out_bam, data)
bam.index(dedup_bam, data["config"])
data["work_bam"] = dedup_bam
elif fastq1 and objectstore.file_exists_or_remote(
fastq1) and fastq1.endswith(".cram"):
data["work_bam"] = fastq1
elif fastq1 is None and "vrn_file" in data:
data["config"]["algorithm"]["variantcaller"] = False
data["work_bam"] = None
elif not fastq1:
raise ValueError("No 'files' specified for input sample: %s" %
dd.get_sample_name(data))
else:
raise ValueError(
"Could not process input file from sample configuration. \n" +
fastq1 + "\nIs the path to the file correct?\n" +
"If it is a fastq file (not pre-aligned BAM or CRAM), "
"is an aligner specified in the input configuration?")
return [[data]]
def process_alignment(data, alt_input=None):
"""Do an alignment of fastq files, preparing a sorted BAM output file.
"""
data = utils.to_single_data(data)
fastq1, fastq2 = dd.get_input_sequence_files(data)
if alt_input:
fastq1, fastq2 = alt_input
config = data["config"]
aligner = config["algorithm"].get("aligner", None)
if fastq1 and objectstore.file_exists_or_remote(fastq1) and aligner:
logger.info("Aligning lane %s with %s aligner" % (data["rgnames"]["lane"], aligner))
data = align_to_sort_bam(fastq1, fastq2, aligner, data)
data = _add_supplemental_bams(data)
elif fastq1 and objectstore.file_exists_or_remote(fastq1) and fastq1.endswith(".bam"):
sort_method = config["algorithm"].get("bam_sort")
bamclean = config["algorithm"].get("bam_clean")
if bamclean is True or bamclean == "picard":
if sort_method and sort_method != "coordinate":
raise ValueError("Cannot specify `bam_clean: picard` with `bam_sort` other than coordinate: %s"
% sort_method)
out_bam = cleanbam.picard_prep(fastq1, data["rgnames"], dd.get_ref_file(data), data["dirs"],
data)
elif bamclean == "fixrg":
out_bam = cleanbam.fixrg(fastq1, data["rgnames"], dd.get_ref_file(data), data["dirs"], data)
elif sort_method:
runner = broad.runner_from_path("picard", config)
out_file = os.path.join(data["dirs"]["work"], "{}-sort.bam".format(
os.path.splitext(os.path.basename(fastq1))[0]))
out_bam = runner.run_fn("picard_sort", fastq1, sort_method, out_file)
else:
out_bam = link_bam_file(fastq1, os.path.join(data["dirs"]["work"], "prealign",
data["rgnames"]["sample"]))
bam.index(out_bam, data["config"])
bam.check_header(out_bam, data["rgnames"], dd.get_ref_file(data), data["config"])
dedup_bam = postalign.dedup_bam(out_bam, data)
bam.index(dedup_bam, data["config"])
data["work_bam"] = dedup_bam
elif fastq1 and objectstore.file_exists_or_remote(fastq1) and fastq1.endswith(".cram"):
data["work_bam"] = fastq1
elif fastq1 is None and "vrn_file" in data:
data["config"]["algorithm"]["variantcaller"] = False
data["work_bam"] = None
elif not fastq1:
raise ValueError("No 'files' specified for input sample: %s" % dd.get_sample_name(data))
else:
raise ValueError("Could not process input file from sample configuration. \n" +
fastq1 +
"\nIs the path to the file correct or is empty?\n" +
"If it is a fastq file (not pre-aligned BAM or CRAM), "
"is an aligner specified in the input configuration?")
if data.get("work_bam"):
# Add stable 'align_bam' target to use for retrieving raw alignment
data["align_bam"] = data["work_bam"]
data = _add_hla_files(data)
return [[data]]
def filter_barcodes(data):
# if data was pre-demultiplexed, there is no need to filter the barcodes
if dd.get_demultiplexed(data):
return [[data]]
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
correction = dd.get_cellular_barcode_correction(data)
bc = get_cellular_barcodes(data)
if not bc:
logger.info("No cellular barcodes found, skipping filtering.")
return [[data]]
bc1 = None
bc2 = None
bc3 = None
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
if isinstance(bc, six.string_types):
bc1 = bc
if len(bc) == 1:
bc1 = bc[0]
if len(bc) > 1:
bc1 = bc[0]
bc2 = bc[1]
if len(bc) == 3:
bc3 = bc[2]
out_base = dd.get_sample_name(data) + ".filtered.fq.gz"
out_file = os.path.join(umi_dir, out_base)
if file_exists(out_file):
data["files"] = [out_file]
return [[data]]
ncores = dd.get_num_cores(data)
cmd = "{umis} cb_filter --cores {ncores} "
if bc1:
cmd += "--bc1 {bc1} "
if correction:
cmd += "--nedit {correction} "
if bc2:
cmd += "--bc2 {bc2} "
if bc3:
cmd += "--bc3 {bc3} "
fq1_cmd = "{fq1} "
fq1_cmd = fq1_cmd.format(fq1=fq1)
cmd += "{fq1_cmd} | gzip -c > {tx_out_file}"
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
safe_makedir(sample_dir)
umis = config_utils.get_program("umis", data, default="umis")
with file_transaction(out_file) as tx_out_file:
message = "Filtering by cellular barcode."
do.run(cmd.format(**locals()), message)
data["files"] = [out_file]
return [[data]]
def filter_barcodes(data):
# if data was pre-demultiplexed, there is no need to filter the barcodes
if dd.get_demultiplexed(data):
return [[data]]
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
correction = dd.get_cellular_barcode_correction(data)
bc = get_cellular_barcodes(data)
if not bc:
logger.info("No cellular barcodes found, skipping filtering.")
return [[data]]
bc1 = None
bc2 = None
bc3 = None
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
if isinstance(bc, six.string_types):
bc1 = bc
if len(bc) == 1:
bc1 = bc[0]
if len(bc) > 1:
bc1 = bc[0]
bc2 = bc[1]
if len(bc) == 3:
bc3 = bc[2]
out_base = dd.get_sample_name(data) + ".filtered.fq.gz"
out_file = os.path.join(umi_dir, out_base)
if file_exists(out_file):
data["files"] = [out_file]
return [[data]]
ncores = dd.get_num_cores(data)
cmd = "{umis} cb_filter --cores {ncores} "
if bc1:
cmd += "--bc1 {bc1} "
if correction:
cmd += "--nedit {correction} "
if bc2:
cmd += "--bc2 {bc2} "
if bc3:
cmd += "--bc3 {bc3} "
fq1_cmd = "{fq1} " if not is_gzipped(fq1) else "<(gzip -cd {fq1}) "
fq1_cmd = fq1_cmd.format(fq1=fq1)
cmd += "{fq1_cmd} | gzip > {tx_out_file}"
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
safe_makedir(sample_dir)
umis = config_utils.get_program("umis", data, default="umis")
with file_transaction(out_file) as tx_out_file:
message = "Filtering by cellular barcode."
do.run(cmd.format(**locals()), message)
data["files"] = [out_file]
return [[data]]
def run_sailfish_index(*samples):
fq1, _ = dd.get_input_sequence_files(samples[0][0])
kmer_size = estimate_kmer_size(fq1)
Build = namedtuple('Build', ['build', 'ref', 'gtf'])
builds = {Build(get_build_string(x), dd.get_ref_file(x), dd.get_gtf_file(x))
for x in dd.sample_data_iterator(samples)}
data = samples[0][0]
indexdirs = {}
for build in builds:
indexdirs[build.build] = sailfish_index(build.ref, build.gtf, data,
build.build, kmer_size)
return samples
def run_sailfish_index(*samples):
fq1, _ = dd.get_input_sequence_files(samples[0][0])
kmer_size = estimate_kmer_size(fq1)
Build = namedtuple('Build', ['build', 'ref', 'gtf'])
builds = {Build(get_build_string(x), dd.get_ref_file(x), dd.get_gtf_file(x))
for x in dd.sample_data_iterator(samples)}
data = samples[0][0]
indexdirs = {}
for build in builds:
indexdirs[build.build] = sailfish_index(build.ref, build.gtf, data,
build.build, kmer_size)
return samples
def filter_barcodes(data):
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
correction = dd.get_cellular_barcode_correction(data)
bc = get_cellular_barcodes(data)
if not bc:
return [[data]]
bc1 = None
bc2 = None
bc3 = None
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
if isinstance(bc, basestring):
bc1 = bc
if len(bc) == 1:
bc1 = bc[0]
if len(bc) > 1:
bc1 = bc[0]
bc2 = bc[1]
if len(bc) == 3:
bc3 = bc[2]
out_base = dd.get_sample_name(data) + ".filtered.fq.gz"
out_file = os.path.join(umi_dir, out_base)
if file_exists(out_file):
data["files"] = [out_file]
return [[data]]
ncores = dd.get_num_cores(data)
cmd = "{umis} cb_filter --cores {ncores} "
if bc1:
cmd += "--bc1 {bc1} "
if correction:
cmd += "--nedit {correction} "
if bc2:
cmd += "--bc2 {bc2} "
if bc3:
cmd += "--bc3 {bc3} "
fq1_cmd = "{fq1} " if not is_gzipped(fq1) else "<(gzip -cd {fq1}) "
fq1_cmd = fq1_cmd.format(fq1=fq1)
cmd += "{fq1_cmd} | gzip > {tx_out_file}"
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
safe_makedir(sample_dir)
umis = config_utils.get_program("umis", data, default="umis")
with file_transaction(out_file) as tx_out_file:
message = "Filtering by cellular barcode."
do.run(cmd.format(**locals()), message)
data["files"] = [out_file]
return [[data]]
def run_rapmap_align(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
rapmap_dir = os.path.join(work_dir, "rapmap", samplename)
gtf_file = dd.get_gtf_file(data)
fasta_file = dd.get_ref_file(data)
out_file = rapmap_align(fq1, fq2, rapmap_dir, gtf_file, fasta_file,
"quasi", data)
data = dd.set_transcriptome_bam(data, out_file)
return [[data]]
def run_rapmap_align(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
rapmap_dir = os.path.join(work_dir, "rapmap", samplename)
gtf_file = dd.get_gtf_file(data)
fasta_file = dd.get_ref_file(data)
out_file = rapmap_align(fq1, fq2, rapmap_dir, gtf_file, fasta_file,
"quasi", data)
data = dd.set_transcriptome_bam(data, out_file)
return [[data]]
def prepare_input_data(config):
""" In case of disambiguation, we want to run fusion calling on
the disambiguated reads, which are in the work_bam file.
As EricScript accepts 2 fastq files as input, we need to convert
the .bam to 2 .fq files.
"""
if not dd.get_disambiguate(config):
return dd.get_input_sequence_files(config)
work_bam = dd.get_work_bam(config)
logger.info("Converting disambiguated reads to fastq...")
fq_files = convert_bam_to_fastq(work_bam, dd.get_work_dir(config), None,
None, config)
return fq_files
def barcode_histogram(data):
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
out_file = os.path.join(sample_dir, "cb-histogram.txt")
if file_exists(out_file):
return [[data]]
fq1_cmd = fq1
cmd = "{umis} cb_histogram {fq1_cmd} > {tx_out_file}"
with file_transaction(out_file) as tx_out_file:
message = "Computing cellular barcode counts for %s." % fq1
do.run(cmd.format(**locals()), message)
return [[data]]
def barcode_histogram(data):
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
out_file = os.path.join(sample_dir, "cb-histogram.txt")
if file_exists(out_file):
return [[data]]
fq1_cmd = fq1
cmd = "{umis} cb_histogram {fq1_cmd} > {tx_out_file}"
with file_transaction(out_file) as tx_out_file:
message = "Computing cellular barcode counts for %s." % fq1
do.run(cmd.format(**locals()), message)
return [[data]]
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
# if RSEM set to run, but the aligner didn't create the transcriptome BAM
# file, make one with bwa
if dd.get_rsem(data) and not dd.get_transcriptome_bam(data):
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("RSEM was flagged to run, but the transcriptome BAM file "
"was not found. Aligning to the transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def sailfish_index(gtf_file, ref_file, data, build):
work_dir = dd.get_work_dir(data)
out_dir = os.path.join(work_dir, "sailfish", "index", build)
sailfish = config_utils.get_program("sailfish", data["config"])
num_cores = dd.get_num_cores(data)
gtf_fa = create_combined_fasta(data)
if file_exists(os.path.join(out_dir, "versionInfo.json")):
return out_dir
with file_transaction(data, out_dir) as tx_out_dir:
fq1, _ = dd.get_input_sequence_files(data)
kmersize = pick_kmersize(fq1)
cmd = ("{sailfish} index -p {num_cores} -t {gtf_fa} -o {tx_out_dir} "
"-k {kmersize}")
message = "Creating sailfish index for {gtf_fa} with {kmersize} bp kmers."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def prepare_input_data(config):
""" In case of disambiguation, we want to run fusion calling on
the disambiguated reads, which are in the work_bam file.
As EricScript accepts 2 fastq files as input, we need to convert
the .bam to 2 .fq files.
"""
if not dd.get_disambiguate(config):
return dd.get_input_sequence_files(config)
work_bam = dd.get_work_bam(config)
logger.info("Converting disambiguated reads to fastq...")
fq_files = convert_bam_to_fastq(
work_bam, dd.get_work_dir(config), None, None, config
)
return fq_files
def run_kallisto_singlecell(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
out_file = kallisto_singlecell(fq1, kallisto_dir, gtf_file, fasta_file, data)
data = dd.set_kallisto_quant(data, out_file)
return [[data]]
def run_salmon_reads(data):
data = utils.to_single_data(data)
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "salmon", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
fasta_file = dd.get_ref_file(data)
out_file = salmon_quant_reads(fq1, fq2, salmon_dir, gtf_file, fasta_file, data)
data = dd.set_salmon(data, out_file)
data = dd.set_salmon_dir(data, salmon_dir)
return [[data]]
def run_kallisto_singlecell(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
out_file = kallisto_singlecell(fq1, kallisto_dir, gtf_file, fasta_file, data)
data = dd.set_kallisto_quant(data, out_file)
return [[data]]
def sailfish_index(gtf_file, ref_file, data, build):
work_dir = dd.get_work_dir(data)
out_dir = os.path.join(work_dir, "sailfish", "index", build)
sailfish = config_utils.get_program("sailfish", data["config"])
num_cores = dd.get_num_cores(data)
gtf_fa = create_combined_fasta(data)
if file_exists(os.path.join(out_dir, "versionInfo.json")):
return out_dir
with file_transaction(data, out_dir) as tx_out_dir:
fq1, _ = dd.get_input_sequence_files(data)
kmersize = pick_kmersize(fq1)
cmd = ("{sailfish} index -p {num_cores} -t {gtf_fa} -o {tx_out_dir} "
"-k {kmersize}")
message = "Creating sailfish index for {gtf_fa} with {kmersize} bp kmers."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def run_rapmap_pseudoalign(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
rapmap_dir = os.path.join(work_dir, "rapmap", samplename)
gtf_file = dd.get_gtf_file(data)
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
out_file = rapmap_pseudoalign(fq1, fq2, rapmap_dir, gtf_file, fasta_file, data)
data = dd.set_transcriptome_bam(data, out_file)
return [[data]]
def run_salmon_reads(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "salmon", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
out_file = salmon_quant_reads(fq1, fq2, salmon_dir, gtf_file, fasta_file, data)
data = dd.set_sailfish(data, out_file)
data = dd.set_sailfish_dir(data, salmon_dir)
return [[data]]
def run_salmon_reads(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "salmon", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
out_file = salmon_quant_reads(fq1, fq2, salmon_dir, gtf_file, fasta_file, data)
data = dd.set_sailfish(data, out_file)
data = dd.set_sailfish_dir(data, salmon_dir)
return [[data]]
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
# if RSEM set to run, but the aligner didn't create the transcriptome BAM
# file, make one with bwa
if dd.get_rsem(data) and not dd.get_transcriptome_bam(data):
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info(
"RSEM was flagged to run, but the transcriptome BAM file "
"was not found. Aligning to the transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def barcode_histogram(data):
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
out_file = os.path.join(sample_dir, "cb-histogram.txt")
filtered_out_file = os.path.join(sample_dir, "cb-histogram-filtered.txt")
fq1_cmd = fq1
cmd = "{umis} cb_histogram {fq1_cmd} > {tx_out_file}"
if not file_exists(out_file):
with file_transaction(out_file) as tx_out_file:
message = "Computing cellular barcode counts for %s." % fq1
do.run(cmd.format(**locals()), message)
cutoff = dd.get_minimum_barcode_depth(data)
filter_barcode_histogram(filtered_out_file, out_file, cutoff)
newdata = dd.set_histogram_counts(data, filtered_out_file)
return [[newdata]]
def barcode_histogram(data):
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
out_file = os.path.join(sample_dir, "cb-histogram.txt")
filtered_out_file = os.path.join(sample_dir, "cb-histogram-filtered.txt")
fq1_cmd = fq1
cmd = "{umis} cb_histogram {fq1_cmd} > {tx_out_file}"
if not file_exists(out_file):
with file_transaction(out_file) as tx_out_file:
message = "Computing cellular barcode counts for %s." % fq1
do.run(cmd.format(**locals()), message)
cutoff = dd.get_minimum_barcode_depth(data)
filter_barcode_histogram(filtered_out_file, out_file, cutoff)
newdata = dd.set_histogram_counts(data, filtered_out_file)
return [[newdata]]
def run_kallisto_rnaseq(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
assert fq2, ("We don't support kallisto for single-end reads and fusion "
"calling with pizzly does not accept single end reads.")
out_file = kallisto_rnaseq(fq1, fq2, kallisto_dir, gtf_file, fasta_file, data)
data = dd.set_kallisto_quant(data, out_file)
return [[data]]
def filter_barcodes(data):
fq1 = dd.get_input_sequence_files(data)[0]
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
correction = dd.get_cellular_barcode_correction(data)
bc = dd.get_cellular_barcodes(data)
if not bc:
return [[data]]
bc1 = None
bc2 = None
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
if isinstance(bc, basestring):
bc1 = bc
if len(bc) == 1:
bc1 = bc[0]
if len(bc) == 2:
bc1 = bc[0]
bc2 = bc[1]
out_base = dd.get_sample_name(data) + ".filtered.fq.gz"
out_file = os.path.join(umi_dir, out_base)
if file_exists(out_file):
data["files"] = [out_file]
return [[data]]
ncores = dd.get_num_cores(data)
cmd = "{umis} cb_filter --cores {ncores} "
if bc1:
cmd += "--bc1 {bc1} "
if correction:
cmd += "--nedit {correction} "
if bc2:
cmd += "--bc2 {bc2} "
fq1_cmd = "{fq1} " if not is_gzipped(fq1) else "<(gzip -cd {fq1}) "
fq1_cmd = fq1_cmd.format(fq1=fq1)
cmd += "{fq1_cmd} | gzip > {tx_out_file}"
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
safe_makedir(sample_dir)
umis = config_utils.get_program("umis", data, default="umis")
with file_transaction(out_file) as tx_out_file:
message = "Filtering by cellular barcode."
do.run(cmd.format(**locals()), message)
data["files"] = [out_file]
return [[data]]
def run_kallisto_rnaseq(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
assert fq2, ("bcbio doesn't support kallisto for single-end reads, we can "
"add support for this if you open up an issue about it here: "
"https://github.com/chapmanb/bcbio-nextgen/issues")
out_file = kallisto_rnaseq(fq1, fq2, kallisto_dir, gtf_file, fasta_file, data)
data = dd.set_kallisto_quant(data, out_file)
return [[data]]
def run_kallisto_rnaseq(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename)
gtf_file = dd.get_transcriptome_gtf(data, default=dd.get_gtf_file(data))
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
assert fq2, ("We don't support kallisto for single-end reads and fusion "
"calling with pizzly does not accept single end reads.")
out_file = kallisto_rnaseq(fq1, fq2, kallisto_dir, gtf_file, fasta_file,
data)
data = dd.set_kallisto_quant(data, out_file)
return [[data]]
def counts_spikein(data):
data = utils.to_single_data(data)
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "spikein", samplename)
fasta_file = dd.get_spikein_fasta(data)
if not fasta_file:
return data
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
kmer = 31 if not dd.get_analysis(data).lower() == "smallrna-seq" else 15
fasta_index = _index_spikein(fasta_file, salmon_dir, data, kmer)
out_file = _salmon_quant_reads(fq1, fq2, salmon_dir, fasta_index, data)
data = dd.set_spikein_counts(data, out_file)
return data
def run_kallisto_rnaseq(data):
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
kallisto_dir = os.path.join(work_dir, "kallisto", samplename)
gtf_file = dd.get_gtf_file(data)
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
assert fq2, ("bcbio doesn't support kallisto for single-end reads, we can "
"add support for this if you open up an issue about it here: "
"https://github.com/bcbio/bcbio-nextgen/issues")
out_file = kallisto_rnaseq(fq1, fq2, kallisto_dir, gtf_file, fasta_file, data)
data = dd.set_kallisto_quant(data, out_file)
return [[data]]
def counts_spikein(data):
data = utils.to_single_data(data)
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "spikein", samplename)
fasta_file = dd.get_spikein_fasta(data)
if not fasta_file:
return data
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
kmer = 31 if not dd.get_analysis(data).lower() == "smallrna-seq" else 15
fasta_index = _index_spikein(fasta_file, salmon_dir, data, kmer)
out_file = _salmon_quant_reads(fq1, fq2, salmon_dir, fasta_index, data)
data = dd.set_spikein_counts(data, out_file)
return data
def run_sailfish(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
if not fastq.is_fastq(fq1):
return [[data]]
sailfish_dir = os.path.join(work_dir, "sailfish", samplename)
gtf_file = dd.get_gtf_file(data)
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
stranded = dd.get_strandedness(data).lower()
out_file = sailfish(fq1, fq2, sailfish_dir, gtf_file, fasta_file, stranded, data)
data = dd.set_sailfish(data, out_file)
data = dd.set_sailfish_dir(data, sailfish_dir)
return [[data]]
def run_sailfish(data):
samplename = dd.get_sample_name(data)
files = dd.get_input_sequence_files(data)
work_dir = dd.get_work_dir(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
if not fastq.is_fastq(fq1):
return [[data]]
sailfish_dir = os.path.join(work_dir, "sailfish", samplename)
gtf_file = dd.get_gtf_file(data)
assert file_exists(gtf_file), "%s was not found, exiting." % gtf_file
fasta_file = dd.get_ref_file(data)
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
stranded = dd.get_strandedness(data).lower()
out_file = sailfish(fq1, fq2, sailfish_dir, gtf_file, fasta_file, stranded, data)
data = dd.set_sailfish(data, out_file)
data = dd.set_sailfish_dir(data, sailfish_dir)
return [[data]]
def run_salmon_reads(data):
data = utils.to_single_data(data)
files = dd.get_input_sequence_files(data)
if bam.is_bam(files[0]):
files = fastq.convert_bam_to_fastq(files[0], data["dirs"]["work"],
data, data["dirs"], data["config"])
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "salmon", samplename)
gtf_file = dd.get_gtf_file(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
fasta_file = dd.get_ref_file(data)
out_file = salmon_quant_reads(fq1, fq2, salmon_dir, gtf_file, fasta_file,
data)
data = dd.set_salmon(data, out_file)
data = dd.set_salmon_dir(data, salmon_dir)
return [[data]]
def generate_transcript_counts(data):
"""Generate counts per transcript and per exon from an alignment"""
data["count_file"] = featureCounts.count(data)
if dd.get_fusion_mode(data, False):
oncofuse_file = oncofuse.run(data)
if oncofuse_file:
data = dd.set_oncofuse_file(data, oncofuse_file)
# if RSEM set to run, but the aligner didn't create the transcriptome BAM
# file, make one with bwa
if dd.get_disambiguate(data):
logger.info("RSEM is not supported yet for disambiguation protocols. "
"See https://github.com/chapmanb/bcbio-nextgen/issues/859")
return [[data]]
if dd.get_rsem(data) and not dd.get_transcriptome_bam(data):
file1, file2 = dd.get_input_sequence_files(data)
ref_file = dd.get_ref_file(data)
logger.info("RSEM was flagged to run, but the transcriptome BAM file "
"was not found. Aligning to the transcriptome with bowtie2.")
data = bowtie2.align_transcriptome(file1, file2, ref_file, data)
return [[data]]
def run_salmon_reads(data):
data = utils.to_single_data(data)
files = dd.get_input_sequence_files(data)
if bam.is_bam(files[0]):
files = fastq.convert_bam_to_fastq(files[0], data["dirs"]["work"],
data, data["dirs"], data["config"])
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "salmon", samplename)
gtf_file = dd.get_gtf_file(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
fasta_file = dd.get_ref_file(data)
out_file = salmon_quant_reads(fq1, fq2, salmon_dir, gtf_file, fasta_file, data)
data = dd.set_salmon(data, out_file)
data = dd.set_salmon_dir(data, salmon_dir)
data = dd.set_salmon_fraglen_file(data, _get_fraglen_file(salmon_dir))
return [[data]]
def run_salmon_decoy(data):
data = utils.to_single_data(data)
files = dd.get_input_sequence_files(data)
if bam.is_bam(files[0]):
files = fastq.convert_bam_to_fastq(files[0], data["dirs"]["work"],
data, data["dirs"], data["config"])
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "salmon", samplename)
gtf_file = dd.get_gtf_file(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
index = salmon_decoy_index(gtf_file, data, os.path.dirname(salmon_dir))
out_file = salmon_quant_reads(fq1, fq2, salmon_dir, gtf_file, data, index)
data = dd.set_salmon(data, out_file)
data = dd.set_salmon_dir(data, salmon_dir)
data = dd.set_salmon_fraglen_file(data, _get_fraglen_file(salmon_dir))
data = dd.update_summary_qc(data, "salmon", base=dd.get_salmon_fraglen_file(data))
return [[data]]
def salmon_index(gtf_file, data, out_dir):
out_dir = os.path.join(out_dir, "index", sailfish.get_build_string(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambiguate(data))
salmon = config_utils.get_program("salmon", dd.get_config(data))
num_cores = dd.get_num_cores(data)
if dd.get_transcriptome_fasta(data):
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data)
assert file_exists(gtf_fa), "%s was not found, exiting." % gtf_fa
tmpdir = dd.get_tmp_dir(data)
out_file = os.path.join(out_dir, "versionInfo.json")
if file_exists(out_file):
return out_dir
files = dd.get_input_sequence_files(data)
kmersize = sailfish.pick_kmersize(files[0])
with file_transaction(data, out_dir) as tx_out_dir:
cmd = "{salmon} index --keepDuplicates -k {kmersize} -p {num_cores} -i {tx_out_dir} -t {gtf_fa}"
message = "Creating Salmon index for {gtf_fa} with {kmersize} bp kmers."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def process_alignment(data):
"""Do an alignment of fastq files, preparing a sorted BAM output file.
"""
fastq1, fastq2 = dd.get_input_sequence_files(data)
config = data["config"]
aligner = config["algorithm"].get("aligner", None)
if fastq1 and utils.file_exists_or_remote(fastq1) and aligner:
logger.info("Aligning lane %s with %s aligner" % (data["rgnames"]["lane"], aligner))
data = align_to_sort_bam(fastq1, fastq2, aligner, data)
data = _add_supplemental_bams(data)
elif fastq1 and os.path.exists(fastq1) and fastq1.endswith(".bam"):
sort_method = config["algorithm"].get("bam_sort")
bamclean = config["algorithm"].get("bam_clean")
if bamclean is True or bamclean == "picard":
if sort_method and sort_method != "coordinate":
raise ValueError("Cannot specify `bam_clean: picard` with `bam_sort` other than coordinate: %s"
% sort_method)
out_bam = cleanbam.picard_prep(fastq1, data["rgnames"], data["sam_ref"], data["dirs"],
data)
elif sort_method:
runner = broad.runner_from_config(config)
out_file = os.path.join(data["dirs"]["work"], "{}-sort.bam".format(
os.path.splitext(os.path.basename(fastq1))[0]))
out_bam = runner.run_fn("picard_sort", fastq1, sort_method, out_file)
else:
out_bam = link_bam_file(fastq1, os.path.join(data["dirs"]["work"], "prealign",
data["rgnames"]["sample"]))
bam.check_header(out_bam, data["rgnames"], data["sam_ref"], data["config"])
dedup_bam = postalign.dedup_bam(out_bam, data)
data["work_bam"] = dedup_bam
elif fastq1 and utils.file_exists_or_remote(fastq1) and fastq1.endswith(".cram"):
data["work_bam"] = fastq1
elif fastq1 is None and "vrn_file" in data:
data["config"]["algorithm"]["variantcaller"] = False
data["work_bam"] = None
else:
raise ValueError("Could not process input file: %s" % fastq1)
return [[data]]
def salmon_index(gtf_file, ref_file, data, out_dir):
out_dir = os.path.join(out_dir, "index", sailfish.get_build_string(data))
if dd.get_disambiguate(data):
out_dir = "-".join([out_dir] + dd.get_disambiguate(data))
salmon = config_utils.get_program("salmon", dd.get_config(data))
num_cores = dd.get_num_cores(data)
if dd.get_transcriptome_fasta(data):
gtf_fa = dd.get_transcriptome_fasta(data)
else:
gtf_fa = sailfish.create_combined_fasta(data)
assert file_exists(gtf_fa), "%s was not found, exiting." % gtf_fa
tmpdir = dd.get_tmp_dir(data)
out_file = os.path.join(out_dir, "versionInfo.json")
if file_exists(out_file):
logger.info("Transcriptome index for %s detected, skipping building." % gtf_fa)
return out_dir
files = dd.get_input_sequence_files(data)
kmersize = sailfish.pick_kmersize(files[0])
with file_transaction(data, out_dir) as tx_out_dir:
cmd = "{salmon} index -k {kmersize} -p {num_cores} -i {tx_out_dir} -t {gtf_fa}"
message = "Creating Salmon index for {gtf_fa} with {kmersize} bp kmers."
do.run(cmd.format(**locals()), message.format(**locals()), None)
return out_dir
def counts_spikein(data):
data = utils.to_single_data(data)
samplename = dd.get_sample_name(data)
work_dir = dd.get_work_dir(data)
salmon_dir = os.path.join(work_dir, "spikein", samplename)
fasta_file = dd.get_spikein_fasta(data)
if not fasta_file:
return data
files = dd.get_input_sequence_files(data)
if len(files) == 2:
fq1, fq2 = files
else:
fq1, fq2 = files[0], None
assert file_exists(fasta_file), "%s was not found, exiting." % fasta_file
readlength = fastq.estimate_read_length(fq1)
if readlength % 2 == 0:
readlength -= 1
kmersize = min(readlength, 31)
logger.info("kmersize used for salmon index at spikein quant: %s" % kmersize)
kmersize = kmersize if not dd.get_analysis(data).lower() == "smallrna-seq" else 15
fasta_index = _index_spikein(fasta_file, salmon_dir, data, kmersize)
out_file = _salmon_quant_reads(fq1, fq2, salmon_dir, fasta_index, data)
data = dd.set_spikein_counts(data, out_file)
return data