def run(self, config, config_file, parallel, dirs, samples):
with prun.start(parallel, samples, config, dirs, "trimming") as run_parallel:
samples = run_parallel("trim_lane", samples)
with prun.start(
_wprogs(parallel, ["aligner"], {"tophat": 8, "tophat2": 8, "star": 30}),
samples,
config,
dirs,
"multicore",
multiplier=alignprep.parallel_multiplier(samples),
) as run_parallel:
samples = disambiguate.split(samples)
samples = run_parallel("process_alignment", samples)
samples = disambiguate.resolve(samples, run_parallel)
with prun.start(
_wprogs(parallel, ["samtools", "gatk", "cufflinks"]), samples, config, dirs, "rnaseqcount"
) as run_parallel:
samples = rnaseq.estimate_expression(samples, run_parallel)
# samples = rnaseq.detect_fusion(samples, run_parallel)
combined = combine_count_files([x[0].get("count_file") for x in samples])
organism = utils.get_in(samples[0][0], ("genome_resources", "aliases", "ensembl"), None)
annotated = annotate_combined_count_file(combined, organism)
for x in samples:
x[0]["combined_counts"] = combined
x[0]["annotated_combined_counts"] = annotated
with prun.start(
_wprogs(parallel, ["picard", "fastqc", "rnaseqc"]), samples, config, dirs, "persample"
) as run_parallel:
samples = qcsummary.generate_parallel(samples, run_parallel)
return samples
def estimate_expression(samples, run_parallel):
samples = run_parallel("generate_transcript_counts", samples)
combined = count.combine_count_files([x[0]["count_file"] for x in samples
if "count_file" in x[0]])
gtf_file = get_in(samples[0][0], ('genome_resources', 'rnaseq',
'transcripts'), None)
annotated = count.annotate_combined_count_file(combined, gtf_file)
samples = run_parallel("run_cufflinks", samples)
#gene
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
to_combine = [x[0]["fpkm"] for x in samples if "fpkm" in x[0]]
fpkm_combined = count.combine_count_files(to_combine, fpkm_combined_file)
#isoform
fpkm_isoform_combined_file = os.path.splitext(combined)[0] + ".isoform.fpkm"
to_combine_isoform = [x[0]["fpkm_isoform"] for x in samples if "fpkm_isoform" in x[0]]
fpkm_isoform_combined = count.combine_count_files(to_combine_isoform,
fpkm_isoform_combined_file,
".isoform.fpkm")
for x in samples:
x[0]["combined_counts"] = combined
if annotated:
x[0]["annotated_combined_counts"] = annotated
if fpkm_combined:
x[0]["combined_fpkm"] = fpkm_combined
if fpkm_isoform_combined:
x[0]["combined_fpkm_isoform"] = fpkm_isoform_combined
return samples
def run(self, config, config_file, parallel, dirs, samples):
with prun.start(_wres(parallel, ["picard"]),
samples, config, dirs, "trimming") as run_parallel:
samples = run_parallel("process_lane", samples)
samples = run_parallel("trim_lane", samples)
with prun.start(_wres(parallel, ["aligner"],
ensure_mem={"tophat": 8, "tophat2": 8, "star": 30}),
samples, config, dirs, "multicore",
multiplier=alignprep.parallel_multiplier(samples)) as run_parallel:
samples = disambiguate.split(samples)
samples = run_parallel("process_alignment", samples)
samples = disambiguate.resolve(samples, run_parallel)
with prun.start(_wres(parallel, ["samtools", "cufflinks"]),
samples, config, dirs, "rnaseqcount") as run_parallel:
samples = rnaseq.estimate_expression(samples, run_parallel)
#samples = rnaseq.detect_fusion(samples, run_parallel)
combined = combine_count_files([x[0].get("count_file") for x in samples])
gtf_file = utils.get_in(samples[0][0], ('genome_resources', 'rnaseq',
'transcripts'), None)
annotated = annotate_combined_count_file(combined, gtf_file)
for x in samples:
x[0]["combined_counts"] = combined
if annotated:
x[0]["annotated_combined_counts"] = annotated
with prun.start(_wres(parallel, ["picard", "fastqc", "rnaseqc"]),
samples, config, dirs, "persample") as run_parallel:
samples = qcsummary.generate_parallel(samples, run_parallel)
return samples
def combine_files(samples):
"""
after quantitation, combine the counts/FPKM/TPM/etc into a single table with
all samples
"""
gtf_file = dd.get_gtf_file(samples[0][0], None)
dexseq_gff = dd.get_dexseq_gff(samples[0][0])
# combine featureCount files
count_files = filter_missing([dd.get_count_file(x[0]) for x in samples])
combined = count.combine_count_files(count_files, ext=".counts")
annotated = count.annotate_combined_count_file(combined, gtf_file)
# combine eXpress files
express_counts_combined = combine_express(samples, combined)
# combine Cufflinks files
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
fpkm_files = filter_missing([dd.get_fpkm(x[0]) for x in samples])
if fpkm_files:
fpkm_combined = count.combine_count_files(fpkm_files, fpkm_combined_file)
else:
fpkm_combined = None
fpkm_isoform_combined_file = os.path.splitext(combined)[0] + ".isoform.fpkm"
isoform_files = filter_missing([dd.get_fpkm_isoform(x[0]) for x in samples])
if isoform_files:
fpkm_isoform_combined = count.combine_count_files(isoform_files,
fpkm_isoform_combined_file,
".isoform.fpkm")
else:
fpkm_isoform_combined = None
# combine DEXseq files
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
to_combine_dexseq = filter_missing([dd.get_dexseq_counts(data[0]) for data in samples])
if to_combine_dexseq:
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file, ".dexseq")
dexseq.create_dexseq_annotation(dexseq_gff, dexseq_combined)
else:
dexseq_combined = None
samples = spikein.combine_spikein(samples)
updated_samples = []
for data in dd.sample_data_iterator(samples):
data = dd.set_combined_counts(data, combined)
if annotated:
data = dd.set_annotated_combined_counts(data, annotated)
if fpkm_combined:
data = dd.set_combined_fpkm(data, fpkm_combined)
if fpkm_isoform_combined:
data = dd.set_combined_fpkm_isoform(data, fpkm_isoform_combined)
if express_counts_combined:
data = dd.set_express_counts(data, express_counts_combined['counts'])
data = dd.set_express_tpm(data, express_counts_combined['tpm'])
data = dd.set_express_fpkm(data, express_counts_combined['fpkm'])
data = dd.set_isoform_to_gene(data, express_counts_combined['isoform_to_gene'])
if dexseq_combined:
data = dd.set_dexseq_counts(data, dexseq_combined_file)
updated_samples.append([data])
return updated_samples
def estimate_expression(samples, run_parallel):
samples = run_parallel("generate_transcript_counts", samples)
count_files = filter_missing([dd.get_count_file(x[0]) for x in samples])
combined = count.combine_count_files(count_files)
gtf_file = dd.get_gtf_file(samples[0][0], None)
annotated = count.annotate_combined_count_file(combined, gtf_file)
samples = run_parallel("run_express", samples)
express_counts_combined = combine_express(samples, combined)
samples = run_parallel("run_cufflinks", samples)
#gene
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
fpkm_files = filter_missing([dd.get_fpkm(x[0]) for x in samples])
fpkm_combined = count.combine_count_files(fpkm_files, fpkm_combined_file)
#isoform
fpkm_isoform_combined_file = os.path.splitext(combined)[0] + ".isoform.fpkm"
isoform_files = filter_missing([dd.get_fpkm_isoform(x[0]) for x in samples])
fpkm_isoform_combined = count.combine_count_files(isoform_files,
fpkm_isoform_combined_file,
".isoform.fpkm")
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
to_combine_dexseq = filter_missing([dd.get_dexseq_counts(data[0]) for data in samples])
if to_combine_dexseq:
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file, ".dexseq")
else:
dexseq_combined = None
updated_samples = []
for data in dd.sample_data_iterator(samples):
data = dd.set_combined_counts(data, combined)
if annotated:
data = dd.set_annotated_combined_counts(data, annotated)
if fpkm_combined:
data = dd.set_combined_fpkm(data, fpkm_combined)
if fpkm_isoform_combined:
data = dd.set_combined_fpkm_isoform(data, fpkm_combined)
if express_counts_combined:
data = dd.set_express_counts(data, express_counts_combined['counts'])
data = dd.set_express_tpm(data, express_counts_combined['tpm'])
data = dd.set_express_fpkm(data, express_counts_combined['fpkm'])
if dexseq_combined:
data = dd.set_dexseq_counts(data, dexseq_combined_file)
updated_samples.append([data])
return updated_samples
def run(self, config, config_file, parallel, dirs, samples):
with prun.start(_wres(parallel, ["picard", "AlienTrimmer"]), samples,
config, dirs, "trimming") as run_parallel:
with profile.report("adapter trimming", dirs):
samples = run_parallel("process_lane", samples)
samples = run_parallel("trim_lane", samples)
with prun.start(_wres(parallel, ["aligner"],
ensure_mem={
"tophat": 8,
"tophat2": 8,
"star": 30
}),
samples,
config,
dirs,
"multicore",
multiplier=alignprep.parallel_multiplier(
samples)) as run_parallel:
with profile.report("alignment", dirs):
samples = disambiguate.split(samples)
samples = run_parallel("process_alignment", samples)
with prun.start(_wres(parallel, ["samtools", "cufflinks"]), samples,
config, dirs, "rnaseqcount") as run_parallel:
with profile.report("disambiguation", dirs):
samples = disambiguate.resolve(samples, run_parallel)
with profile.report("estimate expression", dirs):
samples = rnaseq.estimate_expression(samples, run_parallel)
combined = combine_count_files(
[x[0].get("count_file") for x in samples])
gtf_file = utils.get_in(samples[0][0],
('genome_resources', 'rnaseq', 'transcripts'),
None)
annotated = annotate_combined_count_file(combined, gtf_file)
for x in samples:
x[0]["combined_counts"] = combined
if annotated:
x[0]["annotated_combined_counts"] = annotated
with prun.start(_wres(parallel, ["picard", "fastqc", "rnaseqc"]),
samples, config, dirs, "persample") as run_parallel:
with profile.report("quality control", dirs):
samples = qcsummary.generate_parallel(samples, run_parallel)
logger.info("Timing: finished")
return samples
def run(self, config, config_file, run_parallel, parallel, dirs, lane_items):
lane_items = run_parallel("trim_lane", lane_items)
samples = disambiguate.split(lane_items)
samples = run_parallel("process_alignment", samples)
samples = disambiguate.resolve(samples, run_parallel)
samples = rnaseq.estimate_expression(samples, run_parallel)
#samples = rnaseq.detect_fusion(samples, run_parallel)
combined = combine_count_files([x[0].get("count_file") for x in samples])
organism = utils.get_in(samples[0][0], ('genome_resources', 'aliases',
'ensembl'), None)
annotated = annotate_combined_count_file(combined, organism)
for x in samples:
x[0]["combined_counts"] = combined
x[0]["annotated_combined_counts"] = annotated
samples = qcsummary.generate_parallel(samples, run_parallel)
#run_parallel("generate_bigwig", samples, {"programs": ["ucsc_bigwig"]})
return samples
def estimate_expression(samples, run_parallel):
samples = run_parallel("generate_transcript_counts", samples)
combined = count.combine_count_files(
[x[0]["count_file"] for x in samples if "count_file" in x[0]])
gtf_file = dd.get_gtf_file(samples[0][0], None)
annotated = count.annotate_combined_count_file(combined, gtf_file)
samples = run_parallel("run_cufflinks", samples)
#gene
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
to_combine = [x[0]["fpkm"] for x in samples if "fpkm" in x[0]]
fpkm_combined = count.combine_count_files(to_combine, fpkm_combined_file)
#isoform
fpkm_isoform_combined_file = os.path.splitext(
combined)[0] + ".isoform.fpkm"
to_combine_isoform = [
x[0]["fpkm_isoform"] for x in samples if "fpkm_isoform" in x[0]
]
fpkm_isoform_combined = count.combine_count_files(
to_combine_isoform, fpkm_isoform_combined_file, ".isoform.fpkm")
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
to_combine_dexseq = [dd.get_dexseq_counts(data[0]) for data in samples]
to_combine_dexseq = filter(lambda x: x, to_combine_dexseq)
if to_combine_dexseq:
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file,
".dexseq")
else:
dexseq_combined = None
for x in samples:
x[0]["combined_counts"] = combined
if annotated:
x[0]["annotated_combined_counts"] = annotated
if fpkm_combined:
x[0]["combined_fpkm"] = fpkm_combined
if fpkm_isoform_combined:
x[0]["combined_fpkm_isoform"] = fpkm_isoform_combined
if dexseq_combined:
x[0] = dd.set_dexseq_counts(x[0], dexseq_combined_file)
return samples
def estimate_expression(samples, run_parallel):
samples = run_parallel("generate_transcript_counts", samples)
combined = count.combine_count_files([x[0]["count_file"] for x in samples
if "count_file" in x[0]])
gtf_file = dd.get_gtf_file(samples[0][0], None)
annotated = count.annotate_combined_count_file(combined, gtf_file)
samples = run_parallel("run_cufflinks", samples)
#gene
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
to_combine = [x[0]["fpkm"] for x in samples if "fpkm" in x[0]]
fpkm_combined = count.combine_count_files(to_combine, fpkm_combined_file)
#isoform
fpkm_isoform_combined_file = os.path.splitext(combined)[0] + ".isoform.fpkm"
to_combine_isoform = [x[0]["fpkm_isoform"] for x in samples if "fpkm_isoform" in x[0]]
fpkm_isoform_combined = count.combine_count_files(to_combine_isoform,
fpkm_isoform_combined_file,
".isoform.fpkm")
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
to_combine_dexseq = [dd.get_dexseq_counts(data[0]) for data in samples]
to_combine_dexseq = filter(lambda x: x, to_combine_dexseq)
if to_combine_dexseq:
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file, ".dexseq")
else:
dexseq_combined = None
for x in samples:
x[0]["combined_counts"] = combined
if annotated:
x[0]["annotated_combined_counts"] = annotated
if fpkm_combined:
x[0]["combined_fpkm"] = fpkm_combined
if fpkm_isoform_combined:
x[0]["combined_fpkm_isoform"] = fpkm_isoform_combined
if dexseq_combined:
x[0] = dd.set_dexseq_counts(x[0], dexseq_combined_file)
return samples
def combine_files(samples):
"""
after quantitation, combine the counts/FPKM/TPM/etc into a single table with
all samples
"""
data = samples[0][0]
# prefer the supplied transcriptome gtf file
gtf_file = dd.get_transcriptome_gtf(data, None)
if not gtf_file:
gtf_file = dd.get_gtf_file(data, None)
dexseq_gff = dd.get_dexseq_gff(data)
# combine featureCount files
count_files = filter_missing([dd.get_count_file(x[0]) for x in samples])
combined = count.combine_count_files(count_files, ext=".counts")
annotated = count.annotate_combined_count_file(combined, gtf_file)
# add tx2gene file
tx2gene_file = os.path.join(dd.get_work_dir(data), "annotation",
"tx2gene.csv")
if gtf_file:
tx2gene_file = sailfish.create_combined_tx2gene(data)
# combine eXpress files
express_counts_combined = combine_express(samples, combined)
# combine Cufflinks files
fpkm_files = filter_missing([dd.get_fpkm(x[0]) for x in samples])
if fpkm_files:
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
fpkm_combined = count.combine_count_files(fpkm_files,
fpkm_combined_file)
else:
fpkm_combined = None
isoform_files = filter_missing(
[dd.get_fpkm_isoform(x[0]) for x in samples])
if isoform_files:
fpkm_isoform_combined_file = os.path.splitext(
combined)[0] + ".isoform.fpkm"
fpkm_isoform_combined = count.combine_count_files(
isoform_files, fpkm_isoform_combined_file, ".isoform.fpkm")
else:
fpkm_isoform_combined = None
# combine DEXseq files
to_combine_dexseq = filter_missing(
[dd.get_dexseq_counts(data[0]) for data in samples])
if to_combine_dexseq:
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file,
".dexseq")
dexseq.create_dexseq_annotation(dexseq_gff, dexseq_combined)
else:
dexseq_combined = None
samples = spikein.combine_spikein(samples)
updated_samples = []
for data in dd.sample_data_iterator(samples):
if combined:
data = dd.set_combined_counts(data, combined)
if annotated:
data = dd.set_annotated_combined_counts(data, annotated)
if fpkm_combined:
data = dd.set_combined_fpkm(data, fpkm_combined)
if fpkm_isoform_combined:
data = dd.set_combined_fpkm_isoform(data, fpkm_isoform_combined)
if express_counts_combined:
data = dd.set_express_counts(data,
express_counts_combined['counts'])
data = dd.set_express_tpm(data, express_counts_combined['tpm'])
data = dd.set_express_fpkm(data, express_counts_combined['fpkm'])
data = dd.set_isoform_to_gene(
data, express_counts_combined['isoform_to_gene'])
if dexseq_combined:
data = dd.set_dexseq_counts(data, dexseq_combined_file)
if gtf_file:
data = dd.set_tx2gene(data, tx2gene_file)
updated_samples.append([data])
return updated_samples
def combine_files(samples):
"""
after quantitation, combine the counts/FPKM/TPM/etc into a single table with
all samples
"""
gtf_file = dd.get_gtf_file(samples[0][0], None)
dexseq_gff = dd.get_dexseq_gff(samples[0][0])
# combine featureCount files
count_files = filter_missing([dd.get_count_file(x[0]) for x in samples])
combined = count.combine_count_files(count_files, ext=".counts")
annotated = count.annotate_combined_count_file(combined, gtf_file)
# combine eXpress files
express_counts_combined = combine_express(samples, combined)
# combine Cufflinks files
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
fpkm_files = filter_missing([dd.get_fpkm(x[0]) for x in samples])
if fpkm_files:
fpkm_combined = count.combine_count_files(fpkm_files,
fpkm_combined_file)
else:
fpkm_combined = None
fpkm_isoform_combined_file = os.path.splitext(
combined)[0] + ".isoform.fpkm"
isoform_files = filter_missing(
[dd.get_fpkm_isoform(x[0]) for x in samples])
if isoform_files:
fpkm_isoform_combined = count.combine_count_files(
isoform_files, fpkm_isoform_combined_file, ".isoform.fpkm")
else:
fpkm_isoform_combined = None
# combine DEXseq files
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
to_combine_dexseq = filter_missing(
[dd.get_dexseq_counts(data[0]) for data in samples])
if to_combine_dexseq:
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file,
".dexseq")
dexseq.create_dexseq_annotation(dexseq_gff, dexseq_combined)
else:
dexseq_combined = None
samples = spikein.combine_spikein(samples)
updated_samples = []
for data in dd.sample_data_iterator(samples):
data = dd.set_combined_counts(data, combined)
if annotated:
data = dd.set_annotated_combined_counts(data, annotated)
if fpkm_combined:
data = dd.set_combined_fpkm(data, fpkm_combined)
if fpkm_isoform_combined:
data = dd.set_combined_fpkm_isoform(data, fpkm_isoform_combined)
if express_counts_combined:
data = dd.set_express_counts(data,
express_counts_combined['counts'])
data = dd.set_express_tpm(data, express_counts_combined['tpm'])
data = dd.set_express_fpkm(data, express_counts_combined['fpkm'])
data = dd.set_isoform_to_gene(
data, express_counts_combined['isoform_to_gene'])
if dexseq_combined:
data = dd.set_dexseq_counts(data, dexseq_combined_file)
updated_samples.append([data])
return updated_samples
def combine_files(samples):
"""
after quantitation, combine the counts/FPKM/TPM/etc into a single table with
all samples
"""
data = samples[0][0]
# prefer the supplied transcriptome gtf file
gtf_file = dd.get_transcriptome_gtf(data, None)
if not gtf_file:
gtf_file = dd.get_gtf_file(data, None)
dexseq_gff = dd.get_dexseq_gff(data)
# combine featureCount files
count_files = filter_missing([dd.get_count_file(x[0]) for x in samples])
combined = count.combine_count_files(count_files, ext=".counts")
annotated = count.annotate_combined_count_file(combined, gtf_file)
# add tx2gene file
tx2gene_file = os.path.join(dd.get_work_dir(data), "annotation", "tx2gene.csv")
if gtf_file:
tx2gene_file = sailfish.create_combined_tx2gene(data)
# combine eXpress files
express_counts_combined = combine_express(samples, combined)
# combine Cufflinks files
fpkm_files = filter_missing([dd.get_fpkm(x[0]) for x in samples])
if fpkm_files and combined:
fpkm_combined_file = os.path.splitext(combined)[0] + ".fpkm"
fpkm_combined = count.combine_count_files(fpkm_files, fpkm_combined_file)
else:
fpkm_combined = None
isoform_files = filter_missing([dd.get_fpkm_isoform(x[0]) for x in samples])
if isoform_files and combined:
fpkm_isoform_combined_file = os.path.splitext(combined)[0] + ".isoform.fpkm"
fpkm_isoform_combined = count.combine_count_files(isoform_files,
fpkm_isoform_combined_file,
".isoform.fpkm")
else:
fpkm_isoform_combined = None
# combine DEXseq files
to_combine_dexseq = filter_missing([dd.get_dexseq_counts(data[0]) for data
in samples])
if to_combine_dexseq and combined:
dexseq_combined_file = os.path.splitext(combined)[0] + ".dexseq"
dexseq_combined = count.combine_count_files(to_combine_dexseq,
dexseq_combined_file, ".dexseq")
if dexseq_combined:
dexseq.create_dexseq_annotation(dexseq_gff, dexseq_combined)
else:
dexseq_combined = None
samples = spikein.combine_spikein(samples)
updated_samples = []
for data in dd.sample_data_iterator(samples):
if combined:
data = dd.set_combined_counts(data, combined)
if annotated:
data = dd.set_annotated_combined_counts(data, annotated)
if fpkm_combined:
data = dd.set_combined_fpkm(data, fpkm_combined)
if fpkm_isoform_combined:
data = dd.set_combined_fpkm_isoform(data, fpkm_isoform_combined)
if express_counts_combined:
data = dd.set_express_counts(data, express_counts_combined['counts'])
data = dd.set_express_tpm(data, express_counts_combined['tpm'])
data = dd.set_express_fpkm(data, express_counts_combined['fpkm'])
data = dd.set_isoform_to_gene(data, express_counts_combined['isoform_to_gene'])
if dexseq_combined:
data = dd.set_dexseq_counts(data, dexseq_combined_file)
if gtf_file:
data = dd.set_tx2gene(data, tx2gene_file)
updated_samples.append([data])
return updated_samples
