def trim_additional_merged_contigs(original_contig, merged_contig):
open_merged=open(merged_contig)
fasta_reader=FastaReader(open_merged)
header, sequence = fasta_reader.next()
sp_header = header.split('_')
trim=int(sp_header[-2].strip('os'))
cigar=sp_header[-1]
all_cigar = re.findall('(\d+)([MXDI])', cigar)
for count, type in all_cigar:
if type=="M" or type=="X":
trim+=int(count)
open_merged.close()
if trim>50:
logging.info("trim %s of %s"%(trim, original_contig))
open_contig=open(original_contig)
fasta_reader=FastaReader(open_contig)
header, sequence,fasta_reader.next()
header=header+"trim_%s"%trim
sequence=sequence[trim:]
open_contig.close()
open_contig=open(original_contig, 'w')
if len(sequence)>100:
open_contig.write(">%s\n%s\n"%(header,sequence))
open_contig.close()
def get_fasta_length(fasta_file):
length = 0
with open(fasta_file) as open_fasta:
reader = FastaReader(open_fasta)
header, sequence = reader.next()
length = len(sequence)
return length
def get_fasta_length(fasta_file):
length=0
with open(fasta_file) as open_fasta:
reader = FastaReader(open_fasta)
header,sequence = reader.next()
length=len(sequence)
return length
def __init__(self,
genome_file,
keep_in_memory=True,
keep_until_done=False,
prefix=''):
self.open_genome_file = open_input_file(genome_file)
self.reader = FastaReader(self.open_genome_file)
self.keep_in_memory = keep_in_memory
self.keep_until_done = keep_until_done
self.prefix = prefix
self.all_chr = {}
def correct_contig_file(contig_file, site_name, min_contig_len=101):
dir_name = os.path.dirname(contig_file)
corrected_file = os.path.join(dir_name, 'contigs_corrected.fa')
open_file = open(contig_file)
open_corrected = open(corrected_file, 'w')
fasta_reader = FastaReader(open_file)
nb_seq = 0
max_len = 0
for fasta_record in fasta_reader:
header, sequence = fasta_record
length_A = sum([len(e) for e in re.findall('[A]{7,}', sequence)])
length_C = sum([len(e) for e in re.findall('[C]{7,}', sequence)])
length_G = sum([len(e) for e in re.findall('[G]{7,}', sequence)])
length_T = sum([len(e) for e in re.findall('[T]{7,}', sequence)])
total_repeat = sum([length_A, length_C, length_G, length_T])
if len(sequence
) > min_contig_len and float(total_repeat) / len(sequence) < .5:
nb_seq += 1
if len(sequence) > max_len:
max_len = len(sequence)
header = '%s_pair_%s_length_%s' % (site_name, nb_seq,
len(sequence))
open_corrected.write('>%s\n%s\n' % (header, sequence))
open_corrected.close()
open_file.close()
return (corrected_file, nb_seq, max_len)
def __init__(self,genome_file, keep_in_memory=True, keep_until_done=False, prefix=''):
self.open_genome_file=open_input_file(genome_file)
self.reader=FastaReader(self.open_genome_file)
self.keep_in_memory=keep_in_memory
self.keep_until_done=keep_until_done
self.prefix=prefix
self.all_chr={}
def merge_read1_and_read2_contigs(name, read1_contig, read2_contigs,
output_dir):
open_read2 = open(read2_contigs)
all_fasta2_entries = []
read2_reader = FastaReader(open_read2)
for header, sequence in read2_reader:
all_fasta2_entries.append((header, sequence))
if len(all_fasta2_entries) == 1:
merged_contigs_info = merge_2_contigs(name, read1_contig,
read2_contigs, output_dir)
if merged_contigs_info:
merged_contig_file = os.path.join(output_dir,
'tmp_merged_consensus.fa')
with open(merged_contig_file, 'w') as open_output:
open_output.write('>%s\n%s\n' % merged_contigs_info)
return merged_contig_file
else:
return None
else:
all_fasta2_entries.sort(cmp=compare_fasta_length)
merged_pair = None
remaining = []
for header, sequence in all_fasta2_entries:
cur_pair = os.path.join(output_dir, header + ".fa")
open_pair = open(cur_pair, 'w')
open_pair.write(">%s\n%s\n" % (header, sequence))
open_pair.close()
if not merged_pair:
merged_pair = merge_2_contigs(name, read1_contig, cur_pair,
output_dir)
if not merged_pair:
remaining.append(cur_pair)
else:
results = merge_2_contigs(name + "add", read1_contig, cur_pair,
output_dir)
#TODO: fix this as it doesn't seems to trim and the output file is the same as above but in the mean time disable using False
if False and results:
additional_merged_pair = os.path.join(
output_dir, 'tmp_merged_consensus.fa')
with open(additional_merged_pair, 'w') as open_output:
open_output.write('>%s\n%s\n' % results)
#trim this contig
trim_additional_merged_contigs(cur_pair,
additional_merged_pair)
remaining.append(cur_pair)
merge_file = os.path.join(output_dir, "merged_consensus.fa")
if merged_pair:
merged_pair_file = os.path.join(output_dir,
'tmp_merged_consensus.fa')
with open(merged_pair_file, 'w') as open_output:
open_output.write('>%s\n%s\n' % merged_pair)
tmp = [merged_pair_file]
tmp.extend(remaining)
concatenate_consensus(tmp, merge_file)
return merge_file
else:
return None
def force_merge_consensus(read1_consensus, read2_consensus, output_merge_file):
open_output = open(output_merge_file, 'w')
open_read1 = open(read1_consensus)
open_read2 = open(read2_consensus)
fasta_reader1 = FastaReader(open_read1)
read1_name, read1_sequence = fasta_reader1.next()
open_read1.close()
name = "%s_forced_merged" % read1_name
array = [read1_sequence]
fasta_reader2 = FastaReader(open_read2)
for read2_name, read2_sequence in fasta_reader2:
name = "%s_%s" % (name, read2_name)
array.append("N" * 100)
array.append(read2_sequence)
open_output.write(">%s\n%s\n" % (name, ''.join(array)))
open_read2.close()
open_output.close()
def force_merge_consensus(read1_consensus, read2_consensus, output_merge_file):
open_output = open(output_merge_file,'w')
open_read1 = open(read1_consensus)
open_read2 = open(read2_consensus)
fasta_reader1 = FastaReader(open_read1)
read1_name, read1_sequence = fasta_reader1.next()
open_read1.close()
name="%s_forced_merged"%read1_name
array=[read1_sequence]
fasta_reader2 = FastaReader(open_read2)
for read2_name, read2_sequence in fasta_reader2:
name="%s_%s"%(name,read2_name)
array.append("N"*100)
array.append(read2_sequence)
open_output.write(">%s\n%s\n"%(name, ''.join(array)))
open_read2.close()
open_output.close()
def get_basic_stats(contig_file, all_sites):
open_file = open(contig_file)
fasta_reader = FastaReader(open_file)
for fasta_record in fasta_reader:
header, sequence = fasta_record
match = re.match('(.+)_pair_\d+_length_\d+', header)
site_name = match.group(1)
all_sites[site_name]["number_contig"] += 1
if len(sequence) > all_sites[site_name].get("max_length"):
all_sites[site_name]["max_length"] = len(sequence)
return all_sites
def get_list_of_length(contig_file):
open_file = open(contig_file)
list_length = []
nb_contig = 0
max_length = 0
fasta_reader = FastaReader(open_file)
for fasta_record in fasta_reader:
header, sequence = fasta_record
nb_contig += 1
if len(sequence) > max_length: max_length = len(sequence)
return nb_contig, max_length
def create_sequence_dict_from_contigs_file(contig_file):
print "Read %s" % contig_file
sequence_dictionary = []
all_names = {}
with open(contig_file) as open_file:
for header, sequence in FastaReader(open_file):
sequence_dictionary.append({"SN": header, "LN": len(sequence)})
if all_names.has_key(header):
raise StandardError("Duplicated reference name %s in %s" %
(header, contig_file))
all_names[header] = 1
return sequence_dictionary
def trim_additional_merged_contigs(original_contig, merged_contig):
open_merged = open(merged_contig)
fasta_reader = FastaReader(open_merged)
header, sequence = fasta_reader.next()
sp_header = header.split('_')
trim = int(sp_header[-2].strip('os'))
cigar = sp_header[-1]
all_cigar = re.findall('(\d+)([MXDI])', cigar)
for count, type in all_cigar:
if type == "M" or type == "X":
trim += int(count)
open_merged.close()
if trim > 50:
logging.info("trim %s of %s" % (trim, original_contig))
open_contig = open(original_contig)
fasta_reader = FastaReader(open_contig)
header, sequence, fasta_reader.next()
header = header + "trim_%s" % trim
sequence = sequence[trim:]
open_contig.close()
open_contig = open(original_contig, 'w')
if len(sequence) > 100:
open_contig.write(">%s\n%s\n" % (header, sequence))
open_contig.close()
def analyse_consensus_file(consensus_file):
with open(consensus_file) as open_file:
fasta_reader = FastaReader(open_file)
number_read2_contig = 0
longuest_contigs = 0
is_merged = False
for header, sequence in fasta_reader:
#_merged_os0_71D21M193I or _pair_1_length_452
sp_header = header.split('_')
if len(sp_header) > 2 and sp_header[-3] == 'merged':
type = 'merged'
number_read2_contig += 1
if len(sequence) > longuest_contigs:
longuest_contigs = len(sequence)
is_merged = True
elif len(sp_header) > 3 and sp_header[-4] == 'pair':
type = 'read2_contig'
number_read2_contig += 1
if len(sequence) > longuest_contigs:
longuest_contigs = len(sequence)
else:
type = 'read1_contig'
return number_read2_contig, longuest_contigs, is_merged
class GenomeLoader:
"""Genome Loader take a fasta file and try to find a given chromosome in it.
You can specify so it also keep the all the loaded unused chromosome into memory until they are required (keep_in_memory=True).
You can also specify so it keep the all the loaded chromosome until the object is destroyed (keep_until_done=True).
You can provide a prefix that will be added to the chromsome names if not already there.
"""
def __init__(self,
genome_file,
keep_in_memory=True,
keep_until_done=False,
prefix=''):
self.open_genome_file = open_input_file(genome_file)
self.reader = FastaReader(self.open_genome_file)
self.keep_in_memory = keep_in_memory
self.keep_until_done = keep_until_done
self.prefix = prefix
self.all_chr = {}
def load_all(self):
self.get_chr('***********************************************')
return self.all_chr.keys()
def get_chr(self, chr):
if not chr.startswith(self.prefix):
chr = self.prefix + chr
if self.keep_until_done: #return if loaded already
fasta_seq = self.all_chr.get(chr)
else: #remove if loaded already
fasta_seq = self.all_chr.pop(chr, None)
if fasta_seq:
(header, seq) = fasta_seq
logging.debug('return %s' % header)
return fasta_seq
curr_chr = ''
seq = ''
while not curr_chr == chr:
fasta_seq = self.reader.next()
if fasta_seq:
(header, seq) = fasta_seq
logging.debug('load %s' % header)
curr_chr = header.split()[0]
if not curr_chr.startswith(self.prefix):
curr_chr = self.prefix + curr_chr
if (self.keep_in_memory
and not curr_chr == chr) or self.keep_until_done:
logging.debug('keep %s' % header)
self.all_chr[curr_chr] = fasta_seq
else:
break
return fasta_seq
def next(self):
fasta_seq = None
if len(self.all_chr) > 0:
chr = self.all_chr.keys[0]
if len(self.all_chr) > 0:
chr = self.all_chr.keys()[0]
if self.keep_until_done: #return if loaded already
fasta_seq = self.all_chr.get(chr)
else: #remove if loaded already
fasta_seq = self.all_chr.pop(chr, None)
if not fasta_seq:
fasta_seq = self.reader.next()
if fasta_seq:
(header, seq) = fasta_seq
curr_chr = header.split()[0]
logging.debug('load %s' % header)
if self.keep_until_done:
logging.debug('keep %s' % header)
self.all_chr[curr_chr] = fasta_seq
if fasta_seq:
logging.debug('return %s' % header)
return fasta_seq
return None
def __iter__(self):
return iter(self.next, None)
def __del__(self):
self.open_genome_file.close()
self.all_chr = None
def close(self):
self.__del__()
class GenomeLoader:
"""Genome Loader take a fasta file and try to find a given chromosome in it.
You can specify so it also keep the all the loaded unused chromosome into memory until they are required (keep_in_memory=True).
You can also specify so it keep the all the loaded chromosome until the object is destroyed (keep_until_done=True).
You can provide a prefix that will be added to the chromsome names if not already there.
"""
def __init__(self,genome_file, keep_in_memory=True, keep_until_done=False, prefix=''):
self.open_genome_file=open_input_file(genome_file)
self.reader=FastaReader(self.open_genome_file)
self.keep_in_memory=keep_in_memory
self.keep_until_done=keep_until_done
self.prefix=prefix
self.all_chr={}
def load_all(self):
self.get_chr('***********************************************')
return self.all_chr.keys()
def get_chr(self, chr):
if not chr.startswith(self.prefix):
chr=self.prefix+chr
if self.keep_until_done: #return if loaded already
fasta_seq=self.all_chr.get(chr)
else: #remove if loaded already
fasta_seq=self.all_chr.pop(chr,None)
if fasta_seq:
(header,seq)=fasta_seq
logging.debug('return %s'%header)
return fasta_seq
curr_chr=''
seq=''
while not curr_chr==chr:
fasta_seq=self.reader.next()
if fasta_seq:
(header,seq)=fasta_seq
logging.debug('load %s'%header)
curr_chr=header.split()[0]
if not curr_chr.startswith(self.prefix):
curr_chr=self.prefix+curr_chr
if (self.keep_in_memory and not curr_chr==chr) or self.keep_until_done:
logging.debug('keep %s'%header)
self.all_chr[curr_chr]=fasta_seq
else: break
return fasta_seq
def next(self):
fasta_seq=None
if len(self.all_chr)>0:
chr=self.all_chr.keys[0]
if len(self.all_chr)>0:
chr=self.all_chr.keys()[0]
if self.keep_until_done: #return if loaded already
fasta_seq=self.all_chr.get(chr)
else: #remove if loaded already
fasta_seq=self.all_chr.pop(chr,None)
if not fasta_seq:
fasta_seq=self.reader.next()
if fasta_seq:
(header,seq)=fasta_seq
curr_chr=header.split()[0]
logging.debug('load %s'%header)
if self.keep_until_done:
logging.debug('keep %s'%header)
self.all_chr[curr_chr]=fasta_seq
if fasta_seq:
logging.debug('return %s'%header)
return fasta_seq
return None
def __iter__(self):
return iter(self.next, None)
def __del__(self):
self.open_genome_file.close()
self.all_chr=None
def close(self):
self.__del__()
