def generate_fastq_depending_on_alignment_results(
inSam, outputs, check_multimapFunction=default_check_multimap):
#Function creates new Fastq file from inSam where kept only reads that are
#in accordance with alignment class
#inSam - referemce Sam filename
#outputs - dict {alignment_class : handler}
#where alignment_classes are
#1 = "unique"
#2 = "repeats"
#0 = "unmapped"
#and handler is either file handler or
#None to throw away reads
#
#see analyze_sam_file for check_multimapFunction parameter
startLog(None)
stats = dict([(i, 0) for i in [0, 1, 2]])
for read, category in sam_file_analyzer_generator(inSam,
check_multimapFunction):
stats[category] += 1
if category in outputs.keys():
outputs[category].write("@" + read.query_name + "\n" +
read.query_sequence + "\n+\n")
for q in read.query_qualities:
outputs[category].write(chr(q + 33))
outputs[category].write("\n")
logging.info(
"Sam file " + inSam + " filtered:\n" +
"\n".join([str(i) + ":\t" + str(j) for i, j in stats.iteritems()]))
def __init__(self, R1sam, R2sam, replaceFiles=True, logFileName=None):
#R1sam, R2sam - sam files with left and right reads
#genome - the mirnylab genome instance
#replaceFiles - ensure that output files can be rewritten if exist
self.R1sam = R1sam
self.R2sam = R2sam
self.replaceFiles = replaceFiles
startLog(filename=logFileName)
logging.info("Starting repeat localizer")
random.seed()
def splif_fastq_depending_on_alignment_results(
inFastq,
inSam,
outputs,
check_multimapFunction=default_check_multimap):
#Function creates new Fastq file from inFastq where kept only reads that are
#in accordance with alignment class (based on inSam)
#inFastq - input Fastq filename
#inSam - referemce Sam filename
#outputs - dict
#alignment_class : handler
#where alignment_classes are
#1 = "unique"
#2 = "repeats"
#0 = "unmapped"
#and handler is either file handler or
#None to throw away reads
#
#see analyze_sam_file for check_multimapFunction parameter
startLog(None)
alignment_results = analyze_sam_file(
inSam, check_multimapFunction=check_multimapFunction)
stats = {}
for i in [0, 1, 2]:
if not i in outputs.keys():
outputs[i] = None
else:
stats[i] = 0
with open(inFastq, "r") as fin:
for line in fin:
alignment_class = alignment_results[line[1:].strip().split()[0]]
if outputs[alignment_class] != None:
stats[alignment_class] += 1
outputs[alignment_class].write(line)
outputs[alignment_class].write(fin.next())
outputs[alignment_class].write(fin.next())
outputs[alignment_class].write(fin.next())
else:
fin.next()
fin.next()
fin.next()
logging.info(
"Fastq " + inFastq + " filtered:\n" +
"\n".join([str(i) + ":\t" + str(j) for i, j in stats.iteritems()]))
def __init__(self,
inFastq,
outSam,
enzymeName,
replaceFiles=False,
aligner_function=None,
aligner_parameters=None,
logFileName="HiCmapper.log"):
#inFastq - input fastq file with reads file name
#outSam - file name of the sam file for alignment
#aligner_function - function which call aligning software. Function must accepts agrs inFastq and outSam. Optional
#aligner_parameters - list of additional parameters to be sent to aligner. Paramteres to pass to alignment software
#example: "--very-sencitive -x path-to-file -p 8"
#enzymeName - name of RE enzyme used for Hi-C data
#replaceFiles - replace output and temp files
#logFileName - file name for logs
startLog(logFileName, loglevel=logging.DEBUG)
self.statistics = {} #TODO intrpduce special statistics class
if aligner_function != None:
self.perform_alignmen == aligner_function
self.aligner_parameters = aligner_parameters
assert enzymeName in Bio.Restriction.Restriction_Dictionary.rest_dict.keys(
)
self.enzyme = eval(
"Bio.Restriction." + enzymeName
) #eval is dangerous. Always use check of str that you pass
self.enzymeSeq = self.enzyme.site.upper()
st = self.enzyme.elucidate().find(
"^") #example: EcoRI.elucidate() returns 'G^AATT_C'
end = self.enzyme.elucidate().find("_")
self.enzymeFilledInSeq = (self.enzymeSeq[:end - 1] +
self.enzymeSeq[st:]).upper()
logging.debug("Uing filled-in enzyme seq: " + self.enzymeFilledInSeq)
self.inFastq = inFastq
self.outSam = outSam
self.replace_files = replaceFiles
#hic_check_multimap is a function to check wheather read has multiple alignments
#this function should be defined separately for each aligner
if not hasattr(self, 'hic_check_multimap'):
logging.warning(
"Using default function to check multiple alignments")
self.hic_check_multimap = default_check_multimap
import numpy as np
import sys
sys.path.append("/mnt/storage/home/vsfishman/tmp/HiC_repeats/scripts")
from miscellaneous import startPmDebug,startLog
from NGSseq_utils import *
from Hi_C_mapper import *
startPmDebug()
basedir = "/mnt/storage/home/vsfishman/tmp/HiC_repeats/out/Battulin2015/"
uniqueSam_R1 = basedir+"2500k_bowtie2_R1.sam"
uniqueSam_R2 = basedir+"2500k_bowtie2_R2.sam"
guesResults = basedir+"2500k.unique.fq.trim30.fq.realigned.coords"
guesResults = basedir + "2500k"+".unique.fq"+".trim30.fq"+"R2Untrimmed.paired.coords"
startLog(filename=None)
mapper = Cbowtie2_Hi_C_mapper("","","HindIII",logFileName=None)
reads_pairs = {}
for line in open(guesResults):
line = line.strip().split()
reads_pairs[line[0]] = map(int,line[1:])
dist1 = []
r1notFound = 0
r2notFound = 0
for r1,status in sam_file_analyzer_generator(uniqueSam_R1,check_multimapFunction=mapper.hic_check_multimap):
if status == 1: