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 tagcount(data):
bam = dd.get_transcriptome_bam(data)
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
out_file = os.path.join(sample_dir, dd.get_sample_name(data) + ".mtx")
if file_exists(out_file):
data = dd.set_count_file(data, out_file)
return [[data]]
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
cutoff = dd.get_minimum_barcode_depth(data)
cb_histogram = os.path.join(sample_dir, "cb-histogram.txt")
positional = "--positional" if dd.get_positional_umi(data, False) else ""
gtf_file = dd.get_transcriptome_gtf(data, None)
if gtf_file:
gene_map_file = os.path.join(
dd.get_work_dir(data), "annotation",
os.path.splitext(gtf_file)[0] + "-tx2gene.tsv")
gene_map_file = gtf.tx2genefile(gtf_file, gene_map_file, tsv=True)
gene_map_flag = " --genemap {0} ".format(gene_map_file)
else:
gene_map_flag = ""
message = "Counting alignments of transcripts in %s." % bam
cmd = ("{umis} tagcount {positional} --cb_cutoff {cutoff} --sparse "
"{gene_map_flag}"
"--cb_histogram {cb_histogram} {bam} {tx_out_file}")
out_files = [out_file, out_file + ".rownames", out_file + ".colnames"]
with file_transaction(out_files) as tx_out_files:
tx_out_file = tx_out_files[0]
do.run(cmd.format(**locals()), message)
data = dd.set_count_file(data, out_file)
return [[data]]
def tagcount(data):
bam = dd.get_transcriptome_bam(data)
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
out_prefix = os.path.join(sample_dir, dd.get_sample_name(data))
out_file = out_prefix + ".mtx"
if file_exists(out_file):
data = dd.set_count_file(data, out_file)
return [[data]]
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
cutoff = dd.get_minimum_barcode_depth(data)
cb_histogram = os.path.join(sample_dir, "cb-histogram.txt")
positional = "--positional" if dd.get_positional_umi(data, False) else ""
if use_installed_transcriptome(data):
gtf_file = dd.get_gtf_file(data)
else:
gtf_file = dd.get_transcriptome_gtf(data, None)
if gtf_file:
gene_map_file = os.path.join(
dd.get_work_dir(data), "annotation",
os.path.basename(os.path.splitext(gtf_file)[0]) + "-tx2gene.tsv")
gene_map_file = gtf.tx2genefile(gtf_file, gene_map_file, tsv=True)
gene_map_flag = " --genemap {0} ".format(gene_map_file)
else:
gene_map_flag = ""
message = "Counting alignments of transcripts in %s." % bam
cmd = ("{umis} fasttagcount --cb_cutoff {cutoff} "
"{gene_map_flag} "
"{positional} "
"--cb_histogram {cb_histogram}")
out_files = [out_file, out_file + ".rownames", out_file + ".colnames"]
umi_matrix_file = out_prefix + "-dupes.mtx"
out_files += [
umi_matrix_file, umi_matrix_file + ".rownames",
umi_matrix_file + ".colnames"
]
if has_umi_matrix(data):
umi_matrix_flag = " --umi_matrix {tx_umi_matrix_full} "
else:
umi_matrix_flag = ""
cmd += umi_matrix_flag
cmd += " {bam} {tx_out_file_full}"
with file_transaction(out_files) as tx_out_files:
tx_out_file = tx_out_files[0]
tx_out_file_full = tx_out_file + ".full"
tx_umi_matrix = tx_out_files[3]
tx_umi_matrix_full = tx_out_files[3] + ".full"
do.run(cmd.format(**locals()), message)
cmd = ("{umis} sparse {tx_out_file_full} {tx_out_file}")
message = "Converting %s to sparse format." % tx_out_file_full
do.run(cmd.format(**locals()), message)
if has_umi_matrix(data):
cmd = ("{umis} sparse {tx_umi_matrix_full} {tx_umi_matrix}")
message = "Converting %s to sparse format." % tx_umi_matrix_full
do.run(cmd.format(**locals()), message)
data = dd.set_count_file(data, out_file)
return [[data]]
def tagcount(data):
bam = dd.get_transcriptome_bam(data)
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
out_prefix = os.path.join(sample_dir, dd.get_sample_name(data))
out_file = out_prefix + ".mtx"
if file_exists(out_file):
data = dd.set_count_file(data, out_file)
return [[data]]
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
cutoff = dd.get_minimum_barcode_depth(data)
cb_histogram = os.path.join(sample_dir, "cb-histogram.txt")
positional = "--positional" if dd.get_positional_umi(data, False) else ""
if use_installed_transcriptome(data):
gtf_file = dd.get_gtf_file(data)
else:
gtf_file = dd.get_transcriptome_gtf(data, None)
if gtf_file:
gene_map_file = os.path.join(dd.get_work_dir(data), "annotation",
os.path.splitext(gtf_file)[0] + "-tx2gene.tsv")
gene_map_file = gtf.tx2genefile(gtf_file, gene_map_file, tsv=True)
gene_map_flag = " --genemap {0} ".format(gene_map_file)
else:
gene_map_flag = ""
message = "Counting alignments of transcripts in %s." % bam
cmd = ("{umis} fasttagcount --cb_cutoff {cutoff} "
"{gene_map_flag} "
"{positional} "
"--cb_histogram {cb_histogram}")
out_files = [out_file, out_file + ".rownames", out_file + ".colnames"]
umi_matrix_file = out_prefix + "-dupes.mtx"
out_files += [umi_matrix_file, umi_matrix_file + ".rownames",
umi_matrix_file + ".colnames"]
if has_umi_matrix(data):
umi_matrix_flag = " --umi_matrix {tx_umi_matrix_full} "
else:
umi_matrix_flag = ""
cmd += umi_matrix_flag
cmd += " {bam} {tx_out_file_full}"
with file_transaction(out_files) as tx_out_files:
tx_out_file = tx_out_files[0]
tx_out_file_full = tx_out_file + ".full"
tx_umi_matrix = tx_out_files[3]
tx_umi_matrix_full = tx_out_files[3] + ".full"
do.run(cmd.format(**locals()), message)
cmd = ("{umis} sparse {tx_out_file_full} {tx_out_file}")
message = "Converting %s to sparse format." % tx_out_file_full
do.run(cmd.format(**locals()), message)
if has_umi_matrix(data):
cmd = ("{umis} sparse {tx_umi_matrix_full} {tx_umi_matrix}")
message = "Converting %s to sparse format." % tx_umi_matrix_full
do.run(cmd.format(**locals()), message)
data = dd.set_count_file(data, out_file)
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)
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 tagcount(data):
bam = dd.get_transcriptome_bam(data)
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
out_file = os.path.join(sample_dir, dd.get_sample_name(data) + ".counts")
if file_exists(out_file):
data = dd.set_count_file(data, out_file)
return [[data]]
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
cutoff = dd.get_minimum_barcode_depth(data)
cb_histogram = os.path.join(sample_dir, "cb-histogram.txt")
message = "Counting alignments of transcripts in %s." % bam
cmd = ("{umis} tagcount --positional --cb_cutoff {cutoff} --cb_histogram "
"{cb_histogram} {bam} {tx_out_file}")
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()), message)
data = dd.set_count_file(data, out_file)
return [[data]]
def tagcount(data):
bam = dd.get_transcriptome_bam(data)
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
out_file = os.path.join(sample_dir, dd.get_sample_name(data) + ".counts")
if file_exists(out_file):
data = dd.set_count_file(data, out_file)
return [[data]]
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
cutoff = dd.get_minimum_barcode_depth(data)
cb_histogram = os.path.join(sample_dir, "cb-histogram.txt")
message = "Counting alignments of transcripts in %s." % bam
cmd = ("{umis} tagcount --positional --cb_cutoff {cutoff} --cb_histogram "
"{cb_histogram} {bam} {tx_out_file}")
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()), message)
data = dd.set_count_file(data, out_file)
return [[data]]
def tagcount(data):
bam = dd.get_transcriptome_bam(data)
umi_dir = os.path.join(dd.get_work_dir(data), "umis")
sample_dir = os.path.join(umi_dir, dd.get_sample_name(data))
out_file = os.path.join(sample_dir, dd.get_sample_name(data) + ".mtx")
if file_exists(out_file):
data = dd.set_count_file(data, out_file)
return [[data]]
umis = config_utils.get_program("umis", data, default="umis")
safe_makedir(sample_dir)
cutoff = dd.get_minimum_barcode_depth(data)
cb_histogram = os.path.join(sample_dir, "cb-histogram.txt")
positional = "--positional" if dd.get_positional_umi(data, False) else ""
message = "Counting alignments of transcripts in %s." % bam
cmd = ("{umis} tagcount {positional} --cb_cutoff {cutoff} --sparse "
"--cb_histogram {cb_histogram} {bam} {tx_out_file}")
out_files = [out_file, out_file + ".rownames", out_file + ".colnames"]
with file_transaction(out_files) as tx_out_files:
tx_out_file = tx_out_files[0]
do.run(cmd.format(**locals()), message)
data = dd.set_count_file(data, out_file)
return [[data]]