class BaseSeq(Helper):
def __init__(self, bam, barcodes=None, out=None, ref=None,
rewritten_bam=None,
consensus_reference=None,
consensus_genomes=None,
haplotype_distribution=None,
vcf=None,
chain=None,
crossmap=None,
export=None,
rank=None,
debug=None):
self.bam = bam
self.barcodes = barcodes
self.out = out
self.ref = Reference(ref)
self.rewritten_bam = rewritten_bam
self.rewritten_sorted_bam = rewritten_bam.replace(".bam", ".sorted.bam") if rewritten_bam else None
self.consensus_reference = consensus_reference
self.consensus_genomes = consensus_genomes
self.haplotype_distribution = haplotype_distribution
self.vcf = vcf
self.chain = chain
self.crossmap = crossmap
self.export = export
self.rank = rank
self.debug = int(debug)
def get_barcodes(self):
# simple approach - align, take soft-clipped, and use the arbitrary 20 bases
# intermediate approach - use the seed and extend approach
out = open(self.out, "w")
self.bc = BarCode(self.bam)
self.bc.simple_approach()
for k, v in sorted(self.bc.barcode_to_read.items()):
q = sorted(v)
out.write("%s\t%s\n" % (k, ",".join(q)))
out.close()
def error_correction_barcodes(self):
# start analysis
self.bc = BarCode(self.bam)
sys.stdout.write("[%s] Starting Error Correction Analysis\n" % (self.get_time(),))
# load barcodes
self.bc.load_barcodes(self.barcodes)
sys.stdout.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
# cluster barcodes
self.bc.cluster_barcodes()
sys.stdout.write("[%s] Clustered BarCodes\n" % (self.get_time(),))
def filter_barcodes(self, barcode, export="fastq"):
list_of_ids = []
with open(self.barcodes, "r") as f:
for line in f:
data = line.strip("\r\n").split("\t")
if barcode == data[0]:
list_of_ids = data[1].split(",")
break
self.bc = BarCode(self.bam)
self.bc.filter_and_export(list_of_ids, self.out, export=export)
def sort_and_rewrite_bam(self):
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting Sort and Rewrite BAM\n" % (self.get_time(),))
self.bc.load_barcodes(self.barcodes)
sys.stderr.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
self.bc.sort_and_rewrite_bam(self.rewritten_bam)
pysam.sort("-n", self.rewritten_bam, self.rewritten_sorted_bam.replace(".bam", ""))
sys.stderr.write("[%s] Sort and Rewrite BAM\n" % (self.get_time(),))
def split_bam_by_barcode(self):
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting procedure to split BAM by barcode\n" % (self.get_time(),))
self.bc.split_bam_into_barcodes(self.ref, self.out, self.export)
sys.stderr.write("[%s] Finished splitting BAM by barcode id\n" % (self.get_time(),))
def assemble_consensus_genomes(self):
# build consensus
self.consensus = Consensus(self.rewritten_sorted_bam, self.ref)
sys.stderr.write("[%s] Starting Consensus Building\n" % (self.get_time(),))
self.consensus.build(debug=self.debug)
sys.stderr.write("[%s] Built and Calculated Consensus\n" % (self.get_time(),))
self.consensus.infer_consensus(self.consensus_reference)
sys.stderr.write("[%s] Inferred Consensus\n" % (self.get_time(),))
self.consensus.output_consensus_genomes(self.consensus_genomes)
sys.stderr.write("[%s] Output Consensus Genomes\n" % (self.get_time(),))
self.consensus.output_haplotype_distribution(self.haplotype_distribution)
sys.stderr.write("[%s] Output Haplotype Distribution\n" % (self.get_time(),))
self.quark = Quark(self.ref.sequence)
self.quark.distance_matrix(sorted(self.consensus.freq_distribution.items(),
key=lambda q: q[1],
reverse=True))
self.quark.graph_it()
self.quark.rank_it(self.rank)
sys.stderr.write("[%s] Output Quark Analysis\n" % (self.get_time(),))
self.ovcf = VCF(self.vcf, crossmap=self.crossmap)
self.ovcf.get_variants(self.ref.sequence,
self.consensus.consensus_genomes)
self.ovcf.output_vcf(self.ref.sequence)
sys.stderr.write("[%s] Output VCF\n" % (self.get_time(),))
self.summary_statistics()
sys.stderr.write("[%s] Output Summary Statistics\n" % (self.get_time(),))
self.ochain = Chain(self.chain)
self.ochain.output_chain(self.ref,
self.consensus.inferred_consensus,
self.consensus.inferred_structure)
sys.stderr.write("[%s] Output Chain File\n" % (self.get_time(),))
def summary_statistics(self):
# coverage per genome
# variants per genome
# estimate PCR and sequencing errors
# barcode distribution
f_out = open(self.out, "w")
self.bc = BarCode(self.bam) #TEMP
self.bc.load_barcodes(self.barcodes) #TEMP
self.consensus.output_consensus_coverage(f_out)
self.ovcf.output_variants_distribution(f_out)
self.bc.output_reads_in_barcode_distribution(f_out)
f_out.close()
def run(self):
# Phase 1 - Detection of BarCode
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting BarCode Analysis \n" % (self.get_time(),))
self.bc.simple_approach()
sys.stderr.write("[%s] Analyzed BarCodes \n" % (self.get_time(),))
self.bc.write_barcodes(self.barcodes)
sys.stderr.write("[%s] Wrote BarCodes\n" % (self.get_time(),))
# Phase 2 - Rewrite BAM
sys.stderr.write("[%s] Starting Sort and Rewrite BAM\n" % (self.get_time(),))
self.bc.load_barcodes(self.barcodes)
sys.stderr.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
self.bc.bam.reset()
self.bc.sort_and_rewrite_bam(self.rewritten_bam)
pysam.sort("-n", self.rewritten_bam, self.rewritten_sorted_bam.replace(".bam", ""))
sys.stderr.write("[%s] Sort and Rewrite BAM\n" % (self.get_time(),))
# Phase 3 - Build Consensus
self.consensus = Consensus(self.rewritten_sorted_bam, self.ref)
sys.stderr.write("[%s] Starting Consensus Building\n" % (self.get_time(),))
self.consensus.build()
sys.stderr.write("[%s] Built and Calculated Consensus\n" % (self.get_time(),))
self.consensus.infer_consensus(self.consensus_reference)
sys.stderr.write("[%s] Inferred Consensus\n" % (self.get_time(),))
# Phase 4 - Call Variants and Haplotypes
self.consensus.output_consensus_genomes(self.consensus_genomes)
sys.stderr.write("[%s] Output Consensus Genomes\n" % (self.get_time(),))
self.consensus.output_haplotype_distribution(self.haplotype_distribution)
sys.stderr.write("[%s] Output Haplotype Distribution\n" % (self.get_time(),))
self.ovcf = VCF(self.vcf, crossmap=self.crossmap)
self.ovcf.get_variants(self.ref.sequence,
self.consensus.consensus_genomes)
self.ovcf.output_vcf(self.ref.sequence)
sys.stderr.write("[%s] Output VCF\n" % (self.get_time(),))
# Phase 5 - Summary Statistics and Chain Files
f_out = open(self.out, "w")
self.consensus.output_consensus_coverage(f_out)
self.ovcf.output_variants_distribution(f_out)
self.bc.output_reads_in_barcode_distribution(f_out)
f_out.close()
sys.stderr.write("[%s] Output Summary Statistics\n" % (self.get_time(),))
self.ochain = Chain(self.chain)
self.ochain.output_chain(self.ref,
self.consensus.inferred_consensus,
self.consensus.inferred_structure)
sys.stderr.write("[%s] Output Chain File\n" % (self.get_time(),))
def assemble_genomes(self):
pass
def assemble_genomes_from_fastq(self):
pass