def __init__(self, config, workflow_name="", validate_barcodes=True):
self._config = config
self._workflow_name = workflow_name
self._locus = None
bc_ids = config.get("BARCODE_IDS", "[]")
if isinstance(bc_ids, str):
self._barcode_ids = yaml.load(bc_ids)
else:
self._barcode_ids = bc_ids
try:
with open(config["BARCODES"], "r") as bc_file:
self._all_barcodes = [line.strip()[1:] for line in bc_file if line.startswith(">")]
except KeyError:
self._all_barcodes = []
#raise WorkflowError("Barcode file not specified")
except IOError:
raise WorkflowError("Could not load barcodes")
if validate_barcodes:
if len(self._barcode_ids) and len(self._all_barcodes):
assert all((x in self._all_barcodes for x in self._barcode_ids)), "barcode id not in barcode file"
if len(self._barcode_ids) == 0:
if len(self._all_barcodes) > 0:
self._barcode_ids = self._all_barcodes
else:
raise WorkflowError("No valid barcodes provided")
self._genes = {}
for locus_file in config.get("LOCI", []):
try:
locus = locus_processing.load_locus_yaml(locus_file)
except IOError:
raise WorkflowError("Locus definition file {} does not exist".format(locus_file))
except ValueError:
raise WorkflowError("{} is not a valid locus definition".format(locus_file))
self._genes[locus.name] = locus_file
phase = int(fields[2].split("haplotype")[-1])
return counts[(cluster, phase)]
def count_passes(allele_id, phasing):
if phasing is None:
return 0
counts = phasing.groupby(["cluster", "phase"])["np"].sum()
fields = allele_id.split(".")
cluster = int(fields[1].split("cluster")[-1])
phase = int(fields[2].split("haplotype")[-1])
return counts[(cluster, phase)]
gene = locus_processing.load_locus_yaml(snakemake.input.gene)
def summarize_alleles(barcode):
alleles = load_alleles(
next(f for f in snakemake.input.haplotypes if barcode in f))
vep = load_vep(next(f for f in snakemake.input.vep if barcode in f))
last = load_last(next(f for f in snakemake.input.last if barcode in f))
phasing = load_phasing_summary(
next(f for f in snakemake.input.phasing if barcode in f))
num_alleles = len(alleles)
first = True
allele_info = []
for allele in alleles:
"""
Use PyBedTools to generate a fasta file containing the sequence for a single region
"""
from Bio import SeqIO
import pybedtools
import locus_processing
import yaml
locus = locus_processing.load_locus_yaml(snakemake.input.locus)
try:
with open(snakemake.config["EXPERIMENT"], "r") as infile:
experiment = yaml.safe_load(infile)
start_pos = experiment["targets"][0]["primers"][0]["forward"]["start"]
end_pos = experiment["targets"][0]["primers"][0]["reverse"]["end"]
except (KeyError, IOError):
start_pos = locus.coordinates.start
end_pos = locus.coordinates.end
# create a bed tool for the required region
bed_tool = pybedtools.BedTool([(locus.chromosome.name, start_pos - 1, end_pos)])
# associate the bedtool with the reference genome fasta
bed_tool = bed_tool.sequence(fi=snakemake.input.genome)
# get the sequence and save it
with open(snakemake.output[0], "w") as outfile:
sequence = SeqIO.read(bed_tool.seqfn, "fasta")
SeqIO.write(sequence, outfile, "fasta")