def run_exp(i):
tmpfile_prefix = tmpfile + '.' + str(i)
dna_storage.sample_reads_indel(tmpfile + '.oligos',
tmpfile_prefix + '.reads',
num_reads,
sub_prob=sub_prob,
del_prob=del_prob,
ins_prob=ins_prob,
frac_random_reads=frac_random_reads)
#decode
dna_storage.decode_data(tmpfile_prefix + '.reads',
oligo_length,
tmpfile_prefix + '.decoded',
bin_index_len,
BCH_bits,
LDPC_alpha,
ldpc_code_prefix,
file_size,
0.04,
sync=sync,
sync_pos=sync_pos)
if filecmp.cmp(file_name, tmpfile_prefix + '.decoded'):
correct = 1
else:
correct = 0
os.remove(tmpfile_prefix + '.reads')
os.remove(tmpfile_prefix + '.decoded')
return correct
index_len = (bin_index_len[i]+6*BCH_bits[i])//2
sync_pos = (oligo_len-index_len)//2+index_len
print('sync_pos',sync_pos)
file_size = os.path.getsize(infile_path_prefix+infile_name[i])
print('file_size (in bits):', file_size*8)
num_reads = num_reads_init[i]
# load all reads into list
with open(read_file[i]) as f:
all_reads = f.readlines()
while True:
print('num_reads:',num_reads)
print('reading cost (bases/bit):', (num_reads*oligo_len)/(file_size*8))
success = True
for t in range(num_trials):
sampled_reads = random.sample(all_reads,num_reads)
with open(tmpfile_sampled_reads,'w') as f:
for sampled_read in sampled_reads:
f.write(sampled_read)
dna_storage.remove_barcodes_flexbar(tmpfile_sampled_reads,start_barcodes[i], end_barcodes[i], tmpfile_barcode_removed)
dna_storage.decode_data(tmpfile_barcode_removed,oligo_len,tmpfile_decoded,bin_index_len[i],BCH_bits[i],LDPC_alpha[i],LDPC_path_prefix+LDPC_code[i],file_size,eps=eps[i],sync=sync[i],sync_pos=sync_pos)
if not filecmp.cmp(infile_path_prefix+infile_name[i],tmpfile_decoded):
success = False
os.remove(tmpfile_sampled_reads)
os.remove(tmpfile_decoded)
os.remove(tmpfile_barcode_removed)
if not success:
break
if success:
break
num_reads += 500
while True:
#sample
dna_storage.sample_reads_indel(tmpfile + '.oligos',
tmpfile + '.reads',
num_reads,
sub_prob=sub_prob,
del_prob=del_prob,
ins_prob=ins_prob,
frac_random_reads=frac_random_reads)
#decode
dna_storage.decode_data(tmpfile + '.reads',
oligo_length,
tmpfile + '.decoded',
bin_index_len,
BCH_bits,
LDPC_alpha,
ldpc_code_prefix,
file_size,
0.04,
sync=sync,
sync_pos=sync_pos)
num_it += 1
if filecmp.cmp(file_name, tmpfile + '.decoded'):
num_correct += 1
os.remove(tmpfile + '.decoded')
print('num_it:', num_it)
print('num_correct:', num_correct)
if num_it >= 50 and num_it - num_correct >= 10:
break
error_rate = 1 - num_correct / num_it
print('error rate', error_rate)
from params import *
import sys
sys.path.append('..')
import dna_storage
import filecmp
import os
path_to_oligo_files = '/raid/nanopore/shubham/LDPC_DNA_storage_data/oligo_files_1/'
for i in range(num_files):
index_len = (bin_index_len[i]+6*BCH_bits[i])//2
sync_pos = (oligo_len-index_len)//2+index_len
print('sync_pos',sync_pos)
file_size = os.path.getsize(infile_path_prefix+infile_name[i])
dna_storage.encode_data(infile_path_prefix+infile_name[i],oligo_len,path_to_oligo_files+'reads.'+str(i),BCH_bits[i],LDPC_alpha[i],LDPC_path_prefix+LDPC_code[i],bin_index_len[i],sync = sync[i], sync_pos = sync_pos)
# test decoding to see that there is no strange issue
tmpfile_decoded = 'tmpfile_decoded'
dna_storage.decode_data(path_to_oligo_files+'reads.'+str(i),oligo_len,tmpfile_decoded,bin_index_len[i],BCH_bits[i],LDPC_alpha[i],LDPC_path_prefix+LDPC_code[i],file_size,0.01,sync=sync[i],sync_pos=sync_pos)
assert filecmp.cmp(infile_path_prefix+infile_name[i],tmpfile_decoded)
os.remove(tmpfile_decoded)
with open(path_to_oligo_files+'reads.'+str(i)) as f_reads, open(path_to_oligo_files+'oligos_'+str(i)+'.fa','w') as f_oligos:
for j, line in enumerate(f_reads):
f_oligos.write('>oligos_'+str(i)+'_'+start_barcodes[i]+'_'+end_barcodes[i]+'_'+str(j)+'\n')
f_oligos.write(start_barcodes[i]+line.rstrip('\n')+end_barcodes[i]+'\n')
# NOTE: the files generated are slightly different due to some random padding, but that doesn't affect the decoding