def trim_record(read, trim_at):
new_read = Record()
new_read.name = read.name
new_read.sequence = read.sequence[:trim_at]
if hasattr(read, 'quality'):
new_read.quality = read.quality[:trim_at]
return new_read
def trim_record(read, trim_at):
new_read = Record()
new_read.name = read.name
new_read.sequence = read.sequence[:trim_at]
if hasattr(read, 'quality'):
new_read.quality = read.quality[:trim_at]
return new_read
def trim_record(read, trim_at):
"Utility function: create a new record, trimmed at given location."
new_read = Record()
new_read.name = read.name
new_read.sequence = read.sequence[:trim_at]
if hasattr(read, 'quality'):
new_read.quality = read.quality[:trim_at]
return new_read
def trim_record(read, trim_at):
"Utility function: create a new record, trimmed at given location."
new_read = Record()
new_read.name = read.name
new_read.sequence = read.sequence[:trim_at]
if hasattr(read, 'quality'):
new_read.quality = read.quality[:trim_at]
return new_read
def test_read_type_basic():
# test that basic properties of khmer.Read behave like screed.Record
# Constructing without mandatory arguments should raise an exception
with pytest.raises(TypeError):
Read()
name = "895:1:1:1246:14654 1:N:0:NNNNN"
sequence = "ACGT"
r = Read(name, sequence)
s = Record(dict(name=name, sequence=sequence))
for x in (r, s):
assert x.name == name
assert x.sequence == sequence
assert not hasattr(x, 'quality'), x
assert not hasattr(x, 'annotations'), x
def __init__(self, input, output, ispaired, input1, input2, kmerLen,
lossless, verbose):
self.mutiDict = {} ### kmers for buckets
self.bucketTable = {} ### buckets index and it's reads number
self.sequenceTable = {
} ####bucket as diction index, within bucket include (seq,reverseFlag,indexPos)
self.sequenceTableSave = {}
self.read = {
'sequence': "",
'reverse': "",
'indexPos': "",
'N': "",
'len': "",
'order': ""
}
self.kmerLen = kmerLen
self.bucketIndexLen = kmerLen
# self.blockSize = 1024
self.recordNum = 0
self.onsies = 0 ####the number of bucket which only have one seq
self.leftOnesies = 0 ## the number onesies after re-assigned
self.sequenceN = [
] ### the info of N in read present by (readID,pos,len)
self.reversedNum = 0 ##total reverse number
self.skipZone = 0
self.record = Record() #read record
self.buffer = None
self.sequenceNFlag = False #indicate if the seq include N
self.input = input
self.fileOutputPath = output
self.seqLen = 0
self.fileoutName = {
"index": "",
"cov": "",
"rc": "",
"N": "",
"indexPos": "",
"len": 0,
"order": 0
}
self.paired = ispaired
self.input1 = input1
self.input2 = input2
self.verbose = verbose
self.maxKmer = "T" * kmerLen
self.lossless = lossless
self.removeOutputFile()
#ref_loc_to_build.append(ref_file_names_to_loc[os.path.splitext(name)[0]]) # since bbmap stuff changed the file extension
# This uses khmer to merge the contigs and put them in one fasta file
for loc in ref_loc_to_build:
print(os.path.basename(loc))
fid = khmer.ReadParser(loc)
seq = ""
i = 0
for record in fid:
if i == 0:
header = record.name
seq += "NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN"
seq += record.sequence
i += 1
print("there")
record = Record(name=header, sequence=seq)
write_record(record, out_fid)
fid.close()
out_fid.close()
# This relies on using bbmap to do the contig merging, and then will use cat to concatenate them
# ls *.bz2 | xargs -P 4 -I{} ~/Documents/bbmap/./fuse.sh in={} out=Fused/{}
# cd Fused
# ls *.bz2 | xargs -P 4 -I{} bzip2 -d {}
for loc in ref_file_names_to_loc.values(): #ref_loc_to_build:
with open(loc) as infile:
for line in infile:
out_fid.write(line)
out_fid.close()