Using Biojava 5, I am trying to convert a sequence of Amino Acids to a Nucleotide sequence. Any idea how to do this?

  • $\begingroup$ Welcome to SE.bioinformatics. What have you tried so far? $\endgroup$
    – user438383
    Sep 14, 2020 at 15:24
  • $\begingroup$ Just looked in BioJava documentation - I don't think they have anything to do this. There is bioinformatics.org/sms2/rev_trans.html, but I need to write code for this - so can't use that tool $\endgroup$ Sep 14, 2020 at 16:46
  • $\begingroup$ "convert a sequence of Amino Acids to a Nucleotide sequence". Like you've mentioned in a comment, this is called "reverse translation". Avoid the word "convert" as much as you can - it is as generic as a word can get in our context. It doesn't look like BioJava has a reverse translate provision - you might need to write some code yourself. $\endgroup$
    – Ram RS
    Sep 14, 2020 at 18:08
  • 3
    $\begingroup$ Does this answer your question? How to translate amino acid sequences to Nucleotide sequences $\endgroup$ Sep 14, 2020 at 20:22
  • $\begingroup$ Thanks Chris for that link. Interesting. I ended up using files from NCBI, that worked. $\endgroup$ Sep 20, 2020 at 3:31

1 Answer 1


Note that unlike the reverse (nucleotide sequence to protein sequence), which is uniquely defined given a reading frame, the mapping from protein sequence to nucleotide sequence is not unique (it is not a function in the mathematical sense). The method you use to select which of the many (exponential in input sequence length) possible sequences to output depends critically on what the question is.

If the goal is to design a synthetic gene that will produce a desired protein, then you need to optimize it for expression in the organism into which the gene will be inserted, subject to the constraint of the encoded amino acid sequence.

If the goal is to search for nucleotide sequences in one or more genomes that could give rise to a certain polypeptide, there is a much better way than to do reverse translation at all. In particular, most putative protein-coding open reading frames in most sequenced genomes have already been run through forward translation code and the expected amino acid sequence encoded in them can be found in databases. So, this problem is reduced to a substring search in these databases.

  • $\begingroup$ Very interesting! Thanks. That is what I ended up doing. Matching up FASTA files to get sequences. $\endgroup$ Sep 20, 2020 at 3:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.