5

I want to convert a list of fasta ( protein sequences) in a .text file into corresponding nucleotide sequences. A Google search gives me result of DNA to protein conversion but not vice versa. Also, I came across How do I find the nucleotide sequence of a protein using Biopython?, but this is what I am not looking for. Is there any possible way to do it using python.Moreover, I would like to solve it using python programming. I am sure there must be some way to do it rather than writing a code from scratch. Thanks!

user3289492
  • 51
  • 1
  • 2

3 Answers3

5

Although there is not a unique nucleotide sequence that translates to a given protein, one can list all the possible DNA sequences that do translate to that protein.

An online tool that does just that is Backtranambig, from EMBOSS. It produces a DNA sequence representing all the nucleotide sequences matching the input protein, using IUPAC ambiguity codes.

PedroSebe
  • 216
  • 1
  • 5
2

DNA Chisel (written in Python) can reverse translate a protein sequence:

import dnachisel
from dnachisel.biotools import reverse_translate

record = dnachisel.load_record("seq.fa")
reverse_translate(str(record.seq))


GGTCATATTTTAAAAATGTTTCCT
Peter
  • 2,634
  • 15
  • 33
1

You can't do this because there is redundancy in the genetic code and the same protein sequences can be encoded by different nucleotide sequences. There are 64 codons and ~20 amino acids, e.g. GCT, GCC, GCA and GCG all encode Alanine.

karel
  • 135
  • 1
  • 8
Chris_Rands
  • 3,948
  • 12
  • 31
  • For the sake of the programming exercise, one could use IUPAC ambiguity codes to reflect the set of possible codons for a given amino-acid. – bli Feb 09 '20 at 10:33
  • @bli Agreed, one could even build a model to predict the underlying codons since there is codon usage bias. For practical purposes I'd do a TBLASTN against NT or similar – Chris_Rands Feb 09 '20 at 12:24