[Biojava-l] Question about NiedlemanWunsch class

[Jose Manuel Duarte]

Hi Alexander

Maybe it would be helpful if you can post your code and also even the 
sequences you are looking at, just copy/paste them into the email text. 
Then it would be possible to have a look and see whether it is a bug or 
a feature...

By the way, what version of Biojava are you using? is it 3.1? or are you 
directly getting the latest snapshots from github?

Cheers

Jose



On 04/11/14 23:30, Andreas Prlic wrote:
> Hi Alexander,
>
> Not sure, perhaps the javadoc is misleading (it comes from a high 
> level interface and the lower level implementations might do something 
> slightly different). Does the alignment image look correct? You might 
> want to test this with two shorter and perhaps quite similar 
> sequences. Also, take a look at the SmithWaterman approach, which does 
> local alignments, rather than global. Perhaps your two sequences are 
> too far apart and the global alignment does not yield good results...
>
> Andreas
>
>
>
> On Mon Nov 03 2014 at 11:01:59 AM Александр Бескровный 
> <aleksandr.beskrovnyj at phystech.edu 
> <mailto:aleksandr.beskrovnyj at phystech.edu>> wrote:
>
>     Hello!
>     I have a problem with aligning the sequences with the help of
>     NiedlemanWunsch class. The problem is following: I found the
>     sequences, where method "getSimilarity" got negative numbers, but
>     the description said: "It returns score as a similarity between
>     0.0 and 1.0"
>     I looked at the output of methods "getMaxScore", "getMinScore" and
>     "getScore", which are involved in calculations of method
>     "getSimilarity", and I have noticed, that "getMinScore" and
>     "getMaxScore" got strange numbers (maxScore was less than minScore
>     or both were negative), but output of "getScore" seemed adequate.
>     I used SimpleGapPenalty and SubstitutionMatrixHelper.getNuc4_4()
>     as a parameters of constructor of NiedlemanWunsch class.
>     According to rules, i can’t attach file with sequences to my
>     letter. But in case it is possible, I would kindly ask you to
>     explain me, how to do it correctly,and then i'll send you the
>     sequences. They are too long to be send like a text in a letter.
>
>     Thanks for your attention!
>
>     -- 
>     Regards,
>     Alexander Beskrovnyy
>     _______________________________________________
>     Biojava-l mailing list  - Biojava-l at mailman.open-bio.org
>     <mailto:Biojava-l at mailman.open-bio.org>
>     http://mailman.open-bio.org/mailman/listinfo/biojava-l
>
>
>
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l at mailman.open-bio.org
> http://mailman.open-bio.org/mailman/listinfo/biojava-l

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.open-bio.org/pipermail/biojava-l/attachments/20141105/a760b263/attachment.html>