在 Python 中，除了使用 str.translate() 之外，替换给定字符串中某些字符的最快方法是什么？

Question

What is the fastest way to replace certain characters in a given string other than using str.translate() ?

Given a sequence that only consists of letters "A", "T", "G", and "C", I want to replace each instance of "A" with "T", "T" with "A", "C" with "G", and "G" with "C". To do this, I used an ascii dictionary map = {65:84,84:65,71:67,67:71} , and do sequence.translate(map) . However, in Python 3.8 this appears to be slow. I saw people mention using byte or bytearray to do this, but I just don't know how to make it work.

It looks like I first need to encode the sequence using sequence.encode('ascii', 'ignore') and then use translate() to do the translation?

Can anybody please help me?

For example,

sequence = 'ATGCGTGCGCGACTTT'
# {'A':'T', 'T':'A', 'C':'G', 'G':'C'}
map_dict = {65:84,84:65,71:67,67:71}
# expect 'TACGCACGCGCTGAAA'
sequence.translate(map_dict)

Answer 1

Assumption here is the sequence is very long, then this should be O(1):

If you maintain an index which contains the position of each letter in the sequence, then you just need to update the index to do bulk replacements.

For example given seq = "AGCTTCGA"

index = {"A": {0, 7}, "G": {1, 6}, "C": {2, 5}, "T": {3, 4}}

and if I understand correctly you want to do a swap:

def swap(index, charA, charB):
    tmp = index[charB]
    index[charB] = index[charA]
    index[charA] = tmp

swap(index, "A", "T")
print(index)
# {'A': {3, 4}, 'G': {1, 6}, 'C': {2, 5}, 'T': {0, 7}}

Answer 2

I am going to assume that you want to just replace any occurrence of a string with another. Replace will not work in this case thank you for pointing this out but use:

for i in string:
   match i:
      case "A": i="T"
      case "T": i="A"
      case "C": i="G"
      case "G": i="C"
   continue