def trim_lane(item):
"""
if trim_reads is set with no trimmer specified, default to B-run trimming
only. if trimmer is set to a supported type, perform that trimming
instead.
"""
to_trim = [x for x in item["files"] if x is not None]
dirs = item["dirs"]
config = item["config"]
# this block is to maintain legacy configuration files
trim_reads = config["algorithm"].get("trim_reads", False)
if not trim_reads:
logger.info("Skipping trimming of %s." % (", ".join(to_trim)))
return item
# swap the default to None if trim_reads gets deprecated
if trim_reads == "low_quality" or trim_reads == "true":
logger.info("Trimming low quality ends from %s."
% (", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
if trim_reads == "read_through":
logger.info("Trimming low quality ends and read through adapter "
"sequence from %s." % (", ".join(to_trim)))
out_files = trim_read_through(to_trim, dirs, config)
else:
logger.info("Trimming low quality ends from %s."
% (", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
item["files"] = out_files
return [[item]]
def trim_lane(item):
"""
if trim_reads is set with no trimmer specified, default to B-run trimming
only. if trimmer is set to a supported type, perform that trimming
instead.
"""
to_trim = [x for x in item["files"] if x is not None]
dirs = item["dirs"]
config = item["config"]
# this block is to maintain legacy configuration files
trim_reads = config["algorithm"].get("trim_reads", False)
if not trim_reads:
logger.info("Skipping trimming of %s." % (", ".join(to_trim)))
return item
# swap the default to None if trim_reads gets deprecated
if trim_reads == "low_quality" or trim_reads == "true":
logger.info("Trimming low quality ends from %s." %
(", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
if trim_reads == "read_through":
logger.info("Trimming low quality ends and read through adapter "
"sequence from %s." % (", ".join(to_trim)))
out_files = trim_read_through(to_trim, dirs, config)
else:
logger.info("Trimming low quality ends from %s." %
(", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
item["files"] = out_files
return [[item]]
def process_lane(lane_items, fc_name, fc_date, dirs, config):
"""Prepare lanes, potentially splitting based on barcodes.
"""
lane_name = "%s_%s_%s" % (lane_items[0]['lane'], fc_date, fc_name)
logger.info("Demulitplexing %s" % lane_name)
full_fastq1, full_fastq2 = get_fastq_files(dirs["fastq"], dirs["work"],
lane_items[0], fc_name, config=config)
bc_files = split_by_barcode(full_fastq1, full_fastq2, lane_items,
lane_name, dirs, config)
out = []
for item in lane_items:
config = _update_config_w_custom(config, item)
# Can specify all barcodes but might not have actual sequences
# Would be nice to have a good way to check this is okay here.
if bc_files.has_key(item["barcode_id"]):
for fastq1, fastq2, lane_ext in _prep_fastq_files(item, bc_files, dirs, config):
cur_lane_name = lane_name
cur_lane_desc = item["description"]
if item.get("name", "") and config["algorithm"].get("include_short_name", True):
cur_lane_desc = "%s : %s" % (item["name"], cur_lane_desc)
if item["barcode_id"] is not None:
cur_lane_name += "_%s" % (item["barcode_id"])
if lane_ext is not None:
cur_lane_name += "_s{0}".format(lane_ext)
if config["algorithm"].get("trim_reads", False):
trim_info = brun_trim_fastq([x for x in [fastq1, fastq2] if x is not None],
dirs, config)
fastq1 = trim_info[0]
if fastq2 is not None:
fastq2 = trim_info[1]
out.append((fastq1, fastq2, item, cur_lane_name, cur_lane_desc,
dirs, config))
return out
def trim_lane(fastq1, fastq2, info, lane_name, lane_desc, dirs, config):
"""
if trim_reads is set with no trimmer specified, default to B-run trimming
only. if trimmer is set to a supported type, perform that trimming
instead.
"""
# this block is to maintain legacy configuration files
if not config["algorithm"].get("trim_reads", False):
return [(fastq1, fastq2, info, lane_name, lane_desc, dirs, config)]
# swap the default to None if trim_reads gets deprecated
trimmer = config["algorithm"].get("trimmer", "low_quality")
to_trim = [x for x in [fastq1, fastq2] if x is not None]
if trimmer == "low_quality":
logger.info("Trimming low quality ends from %s."
% (", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
elif trimmer == "adapter":
logger.info("Trimming low quality ends and adapter sequence "
"from %s." % (", ".join(to_trim)))
out_files = cutadapt_trim(to_trim, dirs, config)
else:
logger.info("Skipping trimming of %s." % (", ".join(to_trim)))
out_files = [fastq1, fastq2]
fastq1 = out_files[0]
if fastq2 is not None:
fastq2 = out_files[1]
return [(fastq1, fastq2, info, lane_name, lane_desc, dirs, config)]
def process_lane(lane_items, fc_name, fc_date, dirs, config):
"""Prepare lanes, potentially splitting based on barcodes.
"""
lane_name = "%s_%s_%s" % (lane_items[0]['lane'], fc_date, fc_name)
full_fastq1, full_fastq2 = get_fastq_files(dirs["fastq"], dirs["work"],
lane_items[0], fc_name, config=config)
# Filter phiX
custom_config = _update_config_w_custom(config, lane_items[0])
if custom_config["algorithm"].get("filter_phix", False):
# If we are starting from demultiplexed material, we will skip a lane-wise screening
# Screening will be performed on a sample basis
if custom_config["algorithm"].get("demultiplexed", False):
logger.warn("Will not filter phix lane-wise on already demultiplexed files. " \
"You will have to specify genomes_filter_out option for each sample")
else:
logger.info("Filtering phiX from %s" % lane_name)
info = {"genomes_filter_out": "spiked_phix", "description": lane_name}
processed = remove_contaminants(full_fastq1, full_fastq2, info, lane_name, info["description"], dirs, custom_config)
(full_fastq1, full_fastq2, _, lane_name) = processed[0][0:4]
logger.info("Demultiplexing %s" % lane_name)
bc_files = split_by_barcode(full_fastq1, full_fastq2, lane_items,
lane_name, dirs, config)
out = []
for item in lane_items:
config = _update_config_w_custom(config, item)
# Can specify all barcodes but might not have actual sequences
# Would be nice to have a good way to check this is okay here.
if item["barcode_id"] in bc_files:
fastq1, fastq2 = bc_files[item["barcode_id"]]
cur_lane_name = lane_name
cur_lane_desc = item["description"]
if item.get("name", "") and config["algorithm"].get("include_short_name", True):
cur_lane_desc = "%s : %s" % (item["name"], cur_lane_desc)
if item["barcode_id"] is not None:
cur_lane_name += "_%s" % (item["barcode_id"])
if config["algorithm"].get("trim_reads", False):
trim_info = brun_trim_fastq([x for x in [fastq1, fastq2] if x is not None],
dirs, config)
fastq1 = trim_info[0]
if fastq2 is not None:
fastq2 = trim_info[1]
out.append((fastq1, fastq2, item, cur_lane_name, cur_lane_desc,
dirs, config))
return out
def trim_lane(fastq1, fastq2, info, lane_name, lane_desc, dirs, config):
"""
if trim_reads is set with no trimmer specified, default to B-run trimming
only. if trimmer is set to a supported type, perform that trimming
instead.
"""
to_trim = [x for x in [fastq1, fastq2] if x is not None]
# this block is to maintain legacy configuration files
trim_reads = config["algorithm"].get("trim_reads", False)
if not trim_reads:
logger.info("Skipping trimming of %s." % (", ".join(to_trim)))
return [(fastq1, fastq2, info, lane_name, lane_desc, dirs, config)]
# swap the default to None if trim_reads gets deprecated
if trim_reads == "low_quality" or trim_reads == "true":
logger.info("Trimming low quality ends from %s." %
(", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
if trim_reads == "read_through":
logger.info("Trimming low quality ends and read through adapter "
"sequence from %s." % (", ".join(to_trim)))
out_files = trim_read_through(to_trim, dirs, config)
else:
logger.info("Trimming low quality ends from %s." %
(", ".join(to_trim)))
out_files = brun_trim_fastq(to_trim, dirs, config)
fastq1 = out_files[0]
if fastq2 is not None:
fastq2 = out_files[1]
return [(fastq1, fastq2, info, lane_name, lane_desc, dirs, config)]
def process_lane(lane_items, fc_name, fc_date, dirs, config):
"""Prepare lanes, potentially splitting based on barcodes.
"""
lane_name = "%s_%s_%s" % (lane_items[0]['lane'], fc_date, fc_name)
logger.info("Demulitplexing %s" % lane_name)
full_fastq1, full_fastq2 = get_fastq_files(dirs["fastq"],
dirs["work"],
lane_items[0],
fc_name,
config=_update_config_w_custom(
config, lane_items[0]))
bc_files = split_by_barcode(full_fastq1, full_fastq2, lane_items,
lane_name, dirs, config)
out = []
for item in lane_items:
config = _update_config_w_custom(config, item)
# Can specify all barcodes but might not have actual sequences
# Would be nice to have a good way to check this is okay here.
if bc_files.has_key(item["barcode_id"]):
for fastq1, fastq2, lane_ext in _prep_fastq_files(
item, bc_files, dirs, config):
cur_lane_name = lane_name
cur_lane_desc = item["description"]
if item.get("name", "") and config["algorithm"].get(
"include_short_name", True):
cur_lane_desc = "%s : %s" % (item["name"], cur_lane_desc)
if item["barcode_id"] is not None:
cur_lane_name += "_%s" % (item["barcode_id"])
if lane_ext is not None:
cur_lane_name += "_s{0}".format(lane_ext)
if config["algorithm"].get("trim_reads", False):
trim_info = brun_trim_fastq(
[x for x in [fastq1, fastq2] if x is not None], dirs,
config)
fastq1 = trim_info[0]
if fastq2 is not None:
fastq2 = trim_info[1]
out.append((fastq1, fastq2, item, cur_lane_name, cur_lane_desc,
dirs, config))
return out
def process_lane(lane_items, fc_name, fc_date, dirs, config):
"""Prepare lanes, potentially splitting based on barcodes.
"""
lane_name = "%s_%s_%s" % (lane_items[0]['lane'], fc_date, fc_name)
full_fastq1, full_fastq2 = get_fastq_files(dirs["fastq"],
dirs["work"],
lane_items[0],
fc_name,
config=config)
# Filter phiX
custom_config = _update_config_w_custom(config, lane_items[0])
if custom_config["algorithm"].get("filter_phix", False):
# If we are starting from demultiplexed material, we will skip a lane-wise screening
# Screening will be performed on a sample basis
if custom_config["algorithm"].get("demultiplexed", False):
logger.warn("Will not filter phix lane-wise on already demultiplexed files. " \
"You will have to specify genomes_filter_out option for each sample")
else:
logger.info("Filtering phiX from %s" % lane_name)
info = {
"genomes_filter_out": "spiked_phix",
"description": lane_name
}
processed = remove_contaminants(full_fastq1, full_fastq2, info,
lane_name, info["description"],
dirs, custom_config)
(full_fastq1, full_fastq2, _, lane_name) = processed[0][0:4]
logger.info("Demultiplexing %s" % lane_name)
bc_files = split_by_barcode(full_fastq1, full_fastq2, lane_items,
lane_name, dirs, config)
out = []
for item in lane_items:
config = _update_config_w_custom(config, item)
# Can specify all barcodes but might not have actual sequences
# Would be nice to have a good way to check this is okay here.
if item["barcode_id"] in bc_files:
fastq1, fastq2 = bc_files[item["barcode_id"]]
cur_lane_name = lane_name
cur_lane_desc = item["description"]
if item.get("name", "") and config["algorithm"].get(
"include_short_name", True):
cur_lane_desc = "%s : %s" % (item["name"], cur_lane_desc)
if item["barcode_id"] is not None:
cur_lane_name += "_%s" % (item["barcode_id"])
if config["algorithm"].get("trim_reads", False):
trim_info = brun_trim_fastq(
[x for x in [fastq1, fastq2] if x is not None], dirs,
config)
fastq1 = trim_info[0]
if fastq2 is not None:
fastq2 = trim_info[1]
out.append((fastq1, fastq2, item, cur_lane_name, cur_lane_desc,
dirs, config))
return out
