[Biopython-dev] [Bug 2783] Using alternative start codons in Bio.Seq translate method/function

bugzilla-daemon at portal.open-bio.org bugzilla-daemon at portal.open-bio.org
Tue Mar 10 20:49:30 UTC 2009


http://bugzilla.open-bio.org/show_bug.cgi?id=2783





------- Comment #4 from biopython-bugzilla at maubp.freeserve.co.uk  2009-03-10 16:49 EST -------
On comment #3, Eric wrote:
> 
> How about require_start? Or require_met, if you don't mind how strange
> it looks as English. The name with_start_codon seems like it would take
> a codon or alternate table as the argument.

I think "require_start" is OK.  Or "require_start_codon".

> I also see two choices being made by using this parameter:
> (1) Check that the sequence starts with a valid start codon, and
> if not, raise an exception;

That is what my patch does.  Plus of course translating the valid start
codon as a methionine.

> (2) Use a set of alternate genetic codes for looking up the initial
> methionine.

I'm unsure what you mean here.  If you mean actually having the
translate method search for the first valid start codon, I am
really not keen on this at all. This is complicated, and verges
on gene/ORF finding, which I specifically wanted to avoid:

Peter wrote in comment #0:
>> This proposal is NOT about an option to have the translate
>> function/method search the sequence for the first valid
>> start codon (either in frame or not).

On comment #3, Eric wrote:
> From the other bug's discussion it seems like there are a number of boolean
> options that could reasonably be used with the translate() method, but adding
> them all as keyword args would clutter up the API. What about using a bitmask
> in Bio.Seq that can be used with translate()? The re module takes a bitmask as
> the last parameter for most functions, for example, and it looks pretty clean
> compared to a series of boolean keyword args.

I agree that there is a risk of confusion with too many arguments.  But I don't
think a bitmask would help - I think it makes it worse!  I'm not saying its a
good thing, but we have lots of functions/methods in Biopython already with
lots of arguments (e.g. the standalone BLAST wrappers, or the Bio.Entrez
functions).  On the other hand, I can't immediately think of a single python
function which uses a bitmask.


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.



More information about the Biopython-dev mailing list