def bam2fastq(input, output):
import pybam
with open(output, 'w') as fastqout:
with open(input, 'rb') as bamin:
for title, seq, qual in pybam.read(
bamin, ['sam_qname', 'sam_seq', 'sam_qual']):
fastqout.write("@%s\n%s\n+\n%s\n" % (title, seq, qual))
def total_phy_coverage():
print("Total Physical Coverage started...\n")
f = open('../wig-tracks/tot_phy_coverage.wig', 'w')
# initialize genome_change variable as a list constituted by 0 with length = genomelength
genome_length = 3079196
genome_change = [0] * genome_length
for alignment in pybam.read("../data/lact_sorted.bam"):
# conversion of flag from integer to binary
flag = bin(int(alignment.sam_flag))
# get start position and tlen value
start_pos = int(alignment.sam_pos1) # 4th column
tlen = int(alignment.sam_tlen) # 9th column
if tlen <= 3000 and flag.endswith('1'):
if tlen > 0:
genome_change[start_pos] += 1
genome_change[start_pos + tlen] -= 1
else:
genome_change[start_pos + tlen + 1] += 1
genome_change[start_pos + 1] -= 1
print("Generating .wig file\n")
# print genomic profile as a wiggle file
f.write("fixedStep chrom=genome start=1 step=1 span=1 \n")
current_coverage = 0
# cycle over all positions of the genome
for position in range(genome_length):
current_coverage += genome_change[position]
f.write(str(current_coverage) + '\n')
f.close()
print("done!")
def Read_alignment(titleBam, dicoInit, lstError):
try:
pathSCjson = os.path.join(dicoInit['pathTmpDir'],
titleBam + "_SC.json")
pathSCfasta = os.path.join(dicoInit['pathTmpDir'], titleBam + ".fasta")
FASTA = open(pathSCfasta, 'w')
# Init dicoBam
dicoBam = {}
for pos in range(1, dicoInit["dicoGbk"]['refLength'] + 1, 1):
dicoBam[pos] = { 'nb_reads_F':0, 'nb_reads_R':0,\
'nb_sc_reads_F':0, 'nb_sc_reads_R':0,\
'nb_sc_fasta_F':0, 'nb_sc_fasta_R':0 }
#***** BROWSE READS & SEARCH SCR *****#
# Switch to downsampled BAM if exist
if dicoInit['dicoBam'][titleBam]['path_downsampling'] != "":
dicoInit['dicoBam'][titleBam]['path'] = dicoInit['dicoBam'][
titleBam]['path_downsampling']
for alignment in pybam.read(dicoInit['dicoBam'][titleBam]['path']):
if alignment.file_chromosomes[
alignment.sam_refID] == dicoInit["dicoBam"][titleBam][
'refName'] and alignment.sam_mapq >= dicoInit[
'minQ'] and not alignment.sam_cigar_string.__contains__(
"H"
): # and not explain_sam_flags(alignment.sam_flag).__contains__("second in pair") and not explain_sam_flags(alignment.sam_flag).__contains__("supplementary"):
#***** RETRIEVE positions tuple & lastMapped infos *****#
positionsLstTuple, lastMappedPos, lastMappedPosRead = cigar_list_to_tuple(
alignment.sam_cigar_list, alignment.sam_pos0)
#***** FORWARD reads *****#
if explain_sam_flags(
alignment.sam_flag) == "" or explain_sam_flags(
alignment.sam_flag).__contains__(
"mate reverse strand"):
# Count reads
for posTuple in positionsLstTuple:
try:
dicoBam[posTuple[1] + 1]['nb_reads_F'] += 1
except:
pass # None case
# Count softclipped (right soft-clipping)
length, operation = alignment.sam_cigar_list[
len(alignment.sam_cigar_list) - 1]
if operation == "S":
dicoBam[lastMappedPos]['nb_sc_reads_F'] += 1
# Write to Fasta (apply filter) / not consider 'N'
if length - alignment.sam_seq[len(
alignment.sam_seq
) - length:].count("N") >= dicoInit[
'SCsize'] and lastMappedPos + length <= dicoInit[
"dicoGbk"]['refLength']:
nb_mapped_part = min(dicoInit["MappedPart"],
lastMappedPosRead)
FASTA.write(
">" + str(lastMappedPos) + "_" +
str(nb_mapped_part) + "_scrF_" +
alignment.sam_qname + "\n" +
alignment.sam_seq[len(alignment.sam_seq) -
length - nb_mapped_part:] +
"\n")
dicoBam[lastMappedPos]['nb_sc_fasta_F'] += 1
#***** REVERSE reads *****#
else:
# Count reads
for posTuple in positionsLstTuple:
try:
dicoBam[posTuple[1] + 1]['nb_reads_R'] += 1
except:
pass
# Count softclipped (left soft-clipping)
length, operation = alignment.sam_cigar_list[0]
if operation == "S":
dicoBam[alignment.sam_pos0 + 1]['nb_sc_reads_R'] += 1
# Write to Fasta (apply filter)
if length - alignment.sam_seq[0:length].count(
"N"
) >= dicoInit[
'SCsize'] and alignment.sam_pos0 + 1 - length >= 0:
nb_mapped_part = min(dicoInit["MappedPart"],
length)
FASTA.write(">" + str(alignment.sam_pos0 + 1) +
"_" + str(nb_mapped_part) + "_scrR_" +
alignment.sam_qname + "\n" +
alignment.sam_seq[0:length +
nb_mapped_part] +
"\n")
dicoBam[alignment.sam_pos0 +
1]['nb_sc_fasta_R'] += 1
# CLOSE files
FASTA.close()
# WRITE .json results
JSON = open(pathSCjson, 'wb')
JSON.write(json.dumps(dicoBam))
JSON.close()
except:
exc_type, exc_value, exc_traceback = sys.exc_info()
lstError.append("ReadThread \"" + titleBam + "\": " + str(exc_value) +
" (line " + str(exc_traceback.tb_lineno) + ")")
print("debugging...\n")
cont = 0
for row in bam_f:
length = row.sam_tlen
flag = bin(int(row.sam_flag))
print(flag, length, cont)
#################### MAIN ######################
if __name__ == '__main__':
########################## PART 2 ###########################
# load sorted Lactobacillus bam file (using pybam library):
# refer to a BAM file sorted by genomic position!
sorted_bam = pybam.read('../data/lact_sorted.bam')
# 9) Calculate PHYSICAL COVERAGE, creating related wig file
# phy_coverage(sorted_bam)
# 10) Calculate SEQUENCE COVERAGE, creating related wig file
# sequence_coverage(sorted_bam)
# 11) Calculate INSERT STATS
# get_genome_stats(sorted_bam)
# 12) Calculate AVERAGE INSERTS LENGTH, creating related wig file
# avg_inserts_coverage(sorted_bam)
# Saved values from get_genome_stats() function above
avg = 2101.0225496051385
import pybam
bam_data = pybam.read('./pb_467_2_sr_blasr.bam')
bam_rowData = []
for alignment in bam_data:
bam_rowData.append(alignment.sam_seq)
#print alignment.sam_seq
#print alignment.sam_mapq
print bam_rowData[0]