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 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
"""
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_express(samples, combined):
"""Combine tpm, effective counts and fpkm from express results"""
to_combine = [x[0]["eff_counts"] for x in samples if "eff_counts" in x[0]]
if len(to_combine) > 0:
eff_counts_combined_file = os.path.splitext(combined)[0] + ".isoform.eff_counts"
eff_counts_combined = count.combine_count_files(to_combine, eff_counts_combined_file)
to_combine = [x[0]["tpm_counts"] for x in samples if "tpm_counts" in x[0]]
tpm_counts_combined_file = os.path.splitext(combined)[0] + ".isoform.tpm"
tpm_counts_combined = count.combine_count_files(to_combine, tpm_counts_combined_file)
to_combine = [x[0]["fpkm_counts"] for x in samples if "fpkm_counts" in x[0]]
fpkm_counts_combined_file = os.path.splitext(combined)[0] + ".isoform.eff_fpkm"
fpkm_counts_combined = count.combine_count_files(to_combine, fpkm_counts_combined_file)
return {'counts': eff_counts_combined, 'tpm': tpm_counts_combined,
'fpkm': fpkm_counts_combined}
return None
def create_peaktable(samples):
"""create a table of peak counts per sample to use with differential peak calling
"""
data = dd.get_data_from_sample(samples[0])
peakcounts = []
out_dir = os.path.join(dd.get_work_dir(data), "consensus")
out_file = os.path.join(out_dir, "consensus-counts.tsv")
if dd.get_chip_method(data) == "chip":
for data in dd.sample_data_iterator(samples):
peakcounts.append(tz.get_in(("peak_counts"), data))
elif dd.get_chip_method(data) == "atac":
for data in dd.sample_data_iterator(samples):
if bam.is_paired(dd.get_work_bam(data)):
peakcounts.append(tz.get_in(("peak_counts", "NF"), data))
else:
logger.info(f"Creating peak table from full BAM file because "
f"{dd.get_work_bam(data)} is single-ended.")
peakcounts.append(tz.get_in(("peak_counts", "full"), data))
combined_peaks = count.combine_count_files(peakcounts,
out_file,
ext=".counts")
new_data = []
for data in dd.sample_data_iterator(samples):
data = tz.assoc_in(data, ("peak_counts", "peaktable"), combined_peaks)
new_data.append(data)
new_samples = dd.get_samples_from_datalist(new_data)
return new_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 test_dexseq_combine(self):
count_files = test_data.DEXSEQ_COUNT_FILES
test_file = os.path.join(self.out_dir, "dexseq-combined.txt")
out_file = count.combine_count_files(count_files,
out_file=test_file,
ext=".txt")
self.assertTrue(file_exists(out_file))
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)
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 combine_express(samples, combined):
"""Combine tpm, effective counts and fpkm from express results"""
if not combined:
return None
to_combine = [
dd.get_express_counts(x) for x in dd.sample_data_iterator(samples)
if dd.get_express_counts(x)
]
gtf_file = dd.get_gtf_file(samples[0][0])
isoform_to_gene_file = os.path.join(os.path.dirname(combined),
"isoform_to_gene.txt")
isoform_to_gene_file = express.isoform_to_gene_name(
gtf_file, isoform_to_gene_file,
dd.sample_data_iterator(samples).next())
if len(to_combine) > 0:
eff_counts_combined_file = os.path.splitext(
combined)[0] + ".isoform.express_counts"
eff_counts_combined = count.combine_count_files(
to_combine, eff_counts_combined_file, ext=".counts")
to_combine = [
dd.get_express_tpm(x) for x in dd.sample_data_iterator(samples)
if dd.get_express_tpm(x)
]
tpm_counts_combined_file = os.path.splitext(
combined)[0] + ".isoform.express_tpm"
tpm_counts_combined = count.combine_count_files(
to_combine, tpm_counts_combined_file)
to_combine = [
dd.get_express_fpkm(x) for x in dd.sample_data_iterator(samples)
if dd.get_express_fpkm(x)
]
fpkm_counts_combined_file = os.path.splitext(
combined)[0] + ".isoform.express_fpkm"
fpkm_counts_combined = count.combine_count_files(
to_combine, fpkm_counts_combined_file, ext=".fpkm")
return {
'counts': eff_counts_combined,
'tpm': tpm_counts_combined,
'fpkm': fpkm_counts_combined,
'isoform_to_gene': isoform_to_gene_file
}
return {}
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_express(samples, combined):
"""Combine tpm, effective counts and fpkm from express results"""
to_combine = [dd.get_express_counts(x) for x in
dd.sample_data_iterator(samples) if dd.get_express_counts(x)]
gtf_file = dd.get_gtf_file(samples[0][0])
isoform_to_gene_file = os.path.join(os.path.dirname(combined), "isoform_to_gene.txt")
isoform_to_gene_file = express.isoform_to_gene_name(gtf_file, isoform_to_gene_file)
if len(to_combine) > 0:
eff_counts_combined_file = os.path.splitext(combined)[0] + ".isoform.express_counts"
eff_counts_combined = count.combine_count_files(to_combine, eff_counts_combined_file)
to_combine = [dd.get_express_tpm(x) for x in
dd.sample_data_iterator(samples) if dd.get_express_tpm(x)]
tpm_counts_combined_file = os.path.splitext(combined)[0] + ".isoform.express_tpm"
tpm_counts_combined = count.combine_count_files(to_combine, tpm_counts_combined_file)
to_combine = [dd.get_express_fpkm(x) for x in dd.sample_data_iterator(samples)
if dd.get_express_fpkm(x)]
fpkm_counts_combined_file = os.path.splitext(combined)[0] + ".isoform.express_fpkm"
fpkm_counts_combined = count.combine_count_files(to_combine, fpkm_counts_combined_file)
return {'counts': eff_counts_combined, 'tpm': tpm_counts_combined,
'fpkm': fpkm_counts_combined, 'isoform_to_gene': isoform_to_gene_file}
return {}
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 = run_parallel("generate_transcript_counts", samples)
combined = combine_count_files([x[0].get("count_file") for x in samples])
for x in samples:
x[0]["combined_counts"] = combined
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 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", "picard"],
ensure_mem={
"tophat": 8,
"tophat2": 8,
"star": 40
}),
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 create_peaktable(samples):
"""create a table of peak counts per sample to use with differential peak calling
"""
data = dd.get_data_from_sample(samples[0])
peakcounts = []
out_dir = os.path.join(dd.get_work_dir(data), "consensus")
out_file = os.path.join(out_dir, "consensus-counts.tsv")
if dd.get_chip_method(data) == "chip":
for data in dd.sample_data_iterator(samples):
peakcounts.append(tz.get_in(("peak_counts"), data))
elif dd.get_chip_method(data) == "atac":
for data in dd.sample_data_iterator(samples):
peakcounts.append(tz.get_in(("peak_counts", "NF"), data))
combined_peaks = count.combine_count_files(peakcounts,
out_file,
ext=".counts")
new_data = []
for data in dd.sample_data_iterator(samples):
data = tz.assoc_in(data, ("peak_counts", "peaktable"), combined_peaks)
new_data.append(data)
new_samples = dd.get_samples_from_datalist(new_data)
return new_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
"""
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
def test_dexseq_combine(self):
count_files = test_data.DEXSEQ_COUNT_FILES
test_file = os.path.join(self.out_dir, "dexseq-combined.txt")
out_file = count.combine_count_files(count_files, out_file=test_file,
ext=".txt")
self.assertTrue(file_exists(out_file))
