From cb2c89efa186adce1270a0f6195ea5f48b471d4c Mon Sep 17 00:00:00 2001
From: oboulle <olivier.boulle@inria.fr>
Date: Fri, 14 Apr 2023 14:31:25 +0200
Subject: [PATCH] reword ending of abaab conversion, greatly simplified
---
file_to_dna.py | 26 ++++++++++----------------
1 file changed, 10 insertions(+), 16 deletions(-)
diff --git a/file_to_dna.py b/file_to_dna.py
index baab5a3..30cc834 100755
--- a/file_to_dna.py
+++ b/file_to_dna.py
@@ -112,9 +112,13 @@ def encode_file(input_path: str, output_path: str) -> None:
objective is to fit the payload in the lowest number of blocks, but all blocks must be the same size after addition of non coding stuff (primer/buffer/overhang/bsaI)
"""
+ #TODO
binary_string = convert_file_to_bits(input_path) # get the binary string representing the file content
#binary_string = input_path #TODO REMOVE
+ if (len(binary_string) + CHECK_SUM_SIZE) % 8 in [1, 4, 6]: # add a non coding 0 when size of incompatible length for dna conversion
+ binary_string = "0" + binary_string
+
dna_payload_size = bdc.size_dna_from_bit_len_abaab(len(binary_string) + CHECK_SUM_SIZE) # length of the payload after conversion in dna
#print("payload size",str(dna_payload_size))
@@ -132,14 +136,11 @@ def encode_file(input_path: str, output_path: str) -> None:
# estimate the total dna sequence length after addition of non coding parts
total_dna_size = dna_payload_size + dna_for_assembly_size
#print("total_dna_size",str(total_dna_size))
-
- # possibility to add a non coding base at the end of the sequence to make the last block at correct size when it is not possible to do by adding bits of 0 (very rare case)
- add_base_at_end = False
-
+
# round the number of base per blocks, make sure it is above the minimal block size
final_assembly_size = block_number * max(math.ceil(total_dna_size / block_number), min_total_block_size)
- # case when some bits need to be added to increase the number of bases
+ # case when some bits needs to be added to increase the number of bases
if final_assembly_size != total_dna_size:
# calculate the number of bases to add to the payload to get a round number of equal length blocks
@@ -153,12 +154,6 @@ def encode_file(input_path: str, output_path: str) -> None:
binary_string = math.ceil(filler_length) * "0" + binary_string
#print("updated binary size", str(len(binary_string)))
- # rare case where adding 0 can not solve the problem of round blocks
- if dna_payload_needed_size % 5 == 3:
- # need a non coding base at the end
- #print("added non coding base")
- add_base_at_end = True
-
# apply a filter to the binary string -> shuffle the data to avoid long rows of 0 or 1, and avoid rows repetitions
binary_string = binary_string[::-1] # reverse the binary string, because 2 files can have the same start with ziping methods
@@ -172,7 +167,7 @@ def encode_file(input_path: str, output_path: str) -> None:
# convert binaries into dna sequence
sequence = bdc.binary_to_dna_abaab(filtered_binary_string)
- if add_base_at_end: sequence += bdc.bit_to_dna_balance_GC("0", sequence[-1]) # add a non coding base different from the preceding one
+ #if add_base_at_end: sequence += bdc.bit_to_dna_balance_GC("0", sequence[-1]) # add a non coding base different from the preceding one
total_sequence_size = len(sequence)+ dna_for_assembly_size
@@ -189,7 +184,6 @@ def encode_file(input_path: str, output_path: str) -> None:
# split the sequence into blocks of correct size to add the non payload stuff later # start the block count at 1
sub_sequences_dict = {}
-
if block_number == 1:
sub_sequences_dict["1"] = sequence
elif block_number == 2:
@@ -226,7 +220,7 @@ def decode_file(input_path: str, output_path: str) -> None:
sub_sequences_dict = dfr.read_fasta(input_path)
sequence = "".join(sub_sequences_dict.values())
- #sequence = input_path
+ #sequence = input_path #TODO REMOVE
# convert the dna sequence into a binary string
binary_from_dna_string = bdc.dna_to_binary_abaab(sequence)
@@ -283,8 +277,8 @@ if __name__ == '__main__':
continue
print("i=",str(i))
- seq = encode_file("", binary)
- binary_result = decode_file(seq, "toto")
+ seq = encode_file(binary, "")
+ binary_result = decode_file(seq, "")
if binary != binary_result:
#print(binary)
--
GitLab