fasta_recs = []
for peak_fn in peak_fns :
# if --peak-format is auto, figure format out from extension
if opts.peak_format == 'auto' :
fnbase, fnext = os.path.splitext(peak_fn)
if fnext.lower() == '.bed' : # BED file
peak_fmt = 'BED'
elif fnext.lower() == '.xls' : # MACS file
peak_fmt = 'MACS'
else :
warnings.warn('Peak format specified as auto but file extension \
not recognized in file %s, skipping'%peak_fn)
continue
if peak_fmt == 'BED' :
fasta_recs.extend(bed_to_fasta(peak_fn,nib_db,min_header=opts.min_header))
elif peak_fmt == 'MACS' :
fasta_recs.extend(macs_to_fasta(peak_fn,nib_db,min_header=opts.min_header))
# write out foreground to file
if opts.output :
if opts.wrap_width == -1 :
opts.wrap_width = sys.maxint
write_fasta_to_file(dict(fasta_recs),opts.output,linelen=opts.wrap_width)
else :
for header, seq in fasta_recs :
if opts.wrap_width != -1 :
seq = textwrap.fill(seq,opts.wrap_width)
sys.stdout.write('>%s\n%s\n'%(header,seq))
# load up all the fasta records
fasta_recs = {}
for fasta_fn in fasta_fns :
fasta = fasta_to_dict(fasta_fn)
fasta_recs.update(fasta)
# parse --num-seqs argument
if opts.num_seqs.endswith('x') :
num_seq_factor = float(opts.num_seqs[:-1])
num_seqs = int(len(fasta_recs)*num_seq_factor)
else :
try :
num_seqs = int(opts.num_seqs)
except TypeError :
parser.error("Incorrect format of --num-seqs argument, must either be an integer or a factor ending with x, e.g. 2.5x")
# generate the sequences
gen_seqs = rejection_sample_bg(fasta_recs,organism,num_samples=num_seqs,verbose=opts.verbose)
# write out to file
if opts.output :
write_fasta_to_file(gen_seqs,opts.output)
else :
sys.stdout.write(''.join(['>%s\n%s\n'%(k,v) for k,v in gen_seqs.items()]))
if opts.bed :
bed_f = open(opts.bed_output,'w')
bed_f.write(''.join([k.replace(':','\t').replace('-','\t')+'\n' for k in gen_seqs.keys()]))
bed_f.close()
