def load(filepath, db, action=None):
samfile = _load_sam(filepath, db)
sam_data = Sam(db)
chromosome_data = Chromosome(db)
chromosomes = []
for ref in samfile.references:
name = trim_chromosome_name(ref)
c = chromosome_data.get_by_name(name)
c['ref'] = ref
chromosomes.append(c)
filename = os.path.basename(filepath)
sam = sam_data.get_by_filename(filename)
if sam is None:
logging.error('Error : please load "%s" first' % filename)
samId = sam['id']
# cypileup
cypileup.pileup(samfile, chromosomes, samId, db, action)
samfile.close()
def load(filepath, db):
filename = os.path.basename(filepath)
base, ext = os.path.splitext(filepath)
if not re.match('^\.(sam|bam)', ext):
raise UnsupportedFileError('ERROR: Not supported file format')
sam_data = Sam(db)
chr_data = Chromosome(db)
if sam_data.get_by_filename(filename) is not None:
raise AlreadyLoadedError('WARNING: Already loaded "%s"' % filename)
logging.info("Begin to load '%s'" % filename)
# Convert sam to bam
if ext == '.sam':
insam = pysam.Samfile(filepath, 'r')
filepath = base + '.bam'
outbam = pysam.Samfile(filepath, 'wb', template=insam)
for s in insam:
outbam.write(s)
# Create index if not exist
bai = filepath + '.bai'
if not os.path.isfile(bai):
logging.info("Create index '%s'" % os.path.basename(bai))
pysam.index(filepath)
# load sam
samfile = pysam.Samfile(filepath)
sam_data.append(filename,
samfile.header, samfile.lengths,
samfile.mapped,
samfile.nreferences, samfile.references)
# load chromosome
for ref in samfile.references:
chr_data.append(trim_chromosome_name(ref))
def load(filepath, db, action=None):
filename = os.path.basename(filepath)
file_ext = filename.split('.')[-1]
if file_ext != 'bed':
raise UnsupportedFileError('ERROR: Not supported file format')
bed_data = Bed(db)
chr_data = Chromosome(db)
bed_fragment_data = BedFragment(db)
if bed_data.get_by_filename(filename) is not None:
raise AlreadyLoadedError('WARNING: Already loaded "%s"' % filename)
logging.info("Begin to load '%s'" % filename)
# load bed
bedfile = pybed.BedReader(open(filepath, 'r'))
bed_data.append(filename, "", "", 0, 0)
bed = bed_data.get_by_filename(filename)
# load bed fragments
count = 0
for line in bedfile.yield_lines():
b = bedfile.get_line(line)
c_name = b['chrom']
c_name = c_name.replace('Chr', '')
c_name = c_name.replace('chr', '')
c_name = c_name.replace('.', '')
c = chr_data.get_by_name(c_name)
if c is None:
chr_data.append(c_name)
c = chr_data.get_by_name(c_name)
if bed['rgb'] == 0:
rgb = b['itemRgb'].split(',')
if len(rgb) == 3:
[ir, ig, ib] = map(long, b['itemRgb'].split(','))
else:
[ir, ig, ib] = [0, 0, 0]
else:
[ir, ig, ib] = [0, 0, 0]
if b['strand'] == '+':
strand = 0
elif b['strand'] == '-':
strand = 1
else:
strand = -1
bed_fragment_data.append(bed['id'], c['id'],
long(b['chromStart']),
long(b['chromEnd']),
b['name'],
long(b['score']),
strand,
long(b['thickStart']),
long(b['thickEnd']),
ir, ig, ib,
long(b['blockCount']),
b['blockSizes'],
b['blockStarts'])
count += 1
if bedfile.length >= 100 and count % (bedfile.length / 100) == 0:
if action is not None:
act = action()
if isinstance(act, BedLoaderAction):
progress = (count + 1) * 100 / bedfile.length
act(progress)
logging.debug('Loaded %d fragments' % count)
