Pepsin digest (cleavage) does not work using RE?

Question

Aim is to code the theoretical peptic cleavage of protein sequences in Python. The cleavage (cutting) rule for pepsin is: 1234^567. This is a position, ^ stands for the cleavage point.

There are rules that it mustn't have at all and rules where it must have at least one. The rule for mustn't is in position 2, it mustn't be R, K or H. In position 3 and 7, there can't be a P and in position 4, there mustn't be an R.

If it follows "the must not" rules, then it will check "the must" rules so that if any of one them is met, it will cleave. So if there is a P in position 1 or 2, or if there's an L, W, Y or F in position 4 or 5 then, it will be cleaved (use newline).

This is the code I have so far, but it doesn't work yet:

import re
print('INSERT YOUR PEPTIDE') 
X= input(); 
r=re.findall('(?<=[P])(?<=[P,^R,^K,^H])(?<=([^P])(?<=[L,W,F,Y,^R])(?<=[L,W,Y,F])/D(?<=[^P])',X) 
print(r)

My input is:

TTGWMFKGLYQLKSDEARMHNLQRWYVFKVIEFWGINRKDQHDSRYWFSHFCWKLEHGMVPKCFNELLMPPMSTHRSMFNAYSMSFCVQVKEWRGSISRMCDHHQQQCQYWRLMAGGTLI 

How can I make this work?

Answer 1

I am really not sure if I've understood your rules correctly, but I think you want this:

• If all of these are true:

  • position 2: not R, K or H
  • position 3: not P
  • position 4: not R
  • position 7: not P

• And, if at least one of these is true:

  • position 1: P
  • position 1: P
  • position 4: L or W or Y or F
  • position 5: L or W or Y or F

Then cleave.

If so, then you probably don't want to do it using regular expressions. First because it will be hard, and second because it will be absolutely awful to maintain if you need to come back to it. Instead, try something like this:

#!/usr/bin/env python3
# X = input(); 
X = "TTGWMFKGLYQLKSDEARMHNLQRWYVFKVIEFWGINRKDQHDSRYWFSHFCWKLEHGMVPKCFNELLMPPMSTHRSMFNAYSMSFCVQVKEWRGSISRMCDHHQQQCQYWRLMAGGTLI"
seq = list(X)
for i, res in enumerate(seq):
    if i > len(seq) - 7:
        break
    if (
            seq[i+1] in ('R','K','H') or
            seq[i+2] == 'P' or
            seq[i+3] == 'R' or
            seq[i+6] == 'P'
    ):
        continue
    if res == 'P':
        start = i
    elif seq[i+1] == 'P':
        start = i+1
    else:
        continue
    if (
            seq[i+4] in ('L', 'W', 'Y', 'F') or
            seq[i+5] in ('L', 'W', 'Y', 'F')
    ):
        print("%s\n%s\n%s" %
            ("".join(seq[0:start]),
             "".join(seq[start:start+7]),
             "".join(seq[start+7:])
            )
        )

This will produce the following output when run:

$ foo.py
TTGWMFKGLYQLKSDEARMHNLQRWYVFKVIEFWGINRKDQHDSRYWFSHFCWKLEHGMV
PKCFNEL
LMPPMSTHRSMFNAYSMSFCVQVKEWRGSISRMCDHHQQQCQYWRLMAGGTLI

---

If you insist on using regular expressions, you could separate it into two expressions and use something like this:

import re
X = "TTGWMFKGLYQLKSDEARMHNLQRWYVFKVIEFWGINRKDQHDSRYWFSHFCWKLEHGMVPKCFNELLMPPMSTHRSMFNAYSMSFCVQVKEWRGSISRMCDHHQQQCQYWRLMAGGTLI"
re1='P[^RKH][^P][^R][LWYF].[^P]'
re2='P[^P][^R][LWYF].[^P]'
r=re.findall('(?:P[^RKH][^P][^R][LWYF].[^P])|(?:P[^P][^R][LWYF].[^P])',X) 
print(r)

Before you do, however, please consider the wise words of XKCD 208:

Answer 2

Ok I think I almost solve this problem so I still using RE

import re
myprotein = "TTGWMFKGLYQLKSDEARMHNLQRWYVFKVIEFWGINRKDQHDSRYWFSHFCWKLEHGMVPKCFNELLMPPMSTHRSMFNAYSMSFCVQVKEWRGSISRMCDHHQQQCQYWRLMAGGTLI"
my_protein = []
for protein in myprotein:
    myprotein = protein.rstrip('\n')
    my_protein.append(myprotein)
    my_pro = (''.join(my_protein))

   #cleaves sequence    
peptides = re.sub(r'(?<=[^HRK][^P][^R])(?=[FLWY][^P])|(?<=[^HRK][^P][FLWY])(?=.[^P])','\n', my_pro)

print (peptides)

and it produce to

TTG
W
M
F
KG
LY
Q
L
KSDEARMHN
LQRW
YV
F
KVIE
F
W
GINRKDQHDSRY
WF
SH
F
CW
K
L
EHGMVPKC
FNE
L
LMPPMSTHRSMF
NA
Y
SMS
F
CVQVKE
WRGSISRMCDHHQQQCQ
Y
W
RL
MAGGT
LI

but the true answer should be like (checking from https://web.expasy.org/peptide_cutter/ )

TTG
W
M
F
KG
LY
Q
L
KSDEARMHN
LQRW
YV
F
KVIE
F
W
GINRKDQHDSRY
WF
SH
F
CW
K
L
EHGMVPKC
FNE
L
LMPPMSTHRSMF
NA
Y
SMS
F
CVQVKE
WRGSISRMCDHHQQQCQ
Y
W
RL
MAGGT
L
I 

so how can I make this work?