[EMBOSS] Re: [EMBOSS-BUG] water problems

[Lisa Mullan]

Hi!

EMBOSS is scorring it correctly. You have three gaps in thie alignment.
The first gap incurs a penalty of ten for opeing the gap. And then the
subsequent gaps incur penalties of 0.5 each therefore the calculation is:

(10*5)-10-(2*0.5) = 39

I think you got over zealous with your extensions!!

Lisa

Lisa Mullan
MRC Rosalind Franklin Centre for Genomics Research
Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SB, UK
Tel: +44 1223 494500  Fax: +44 1223 494512
E-mail: lmullan at rfcgr.mrc.ac.uk  Web: http://www.rfcgr.mrc.ac.uk

On Wed, 3 Dec 2003, Gila Lithwick wrote:

> Hello,
>
> I wasn't sure which mailing list this e-mail is more appropriate for, so
> I've sent it to both...
>
> I have been using EMBOSS, and noticed the following problems with water:
>
> 1) Water does not give a gap extension penalty for the first gap.  This
> is not how gaps are generally penalized.
> For example, the alignment between tttttcccaaaaa and tttttaaaaa is:
> seq1               1 tttttcccaaaaa     13
>                      |||||   |||||
> seq2               1 ttttt---aaaaa     10
> EMBOSS scores this 39, however the score should be 10*5-3*0.5-10=38.5
> Is this correct?
>
> 2) Since the default gap extension penalty is so low (0.5), water often
> gives very long alignments.  A slightly higher gap penalty (for example
> 1) gives better alignments - maybe it would be better to make this the
> default.
>
> 3) It would be very useful if there was a way to assess significance of
> pairwise alignments, such as the -rand option in gcg.  It is possible,
> of course, to do each step (shuffling one sequence, aligning, and
> calculating the mean and standard deviation) separately.  However, for
> users that don't know how to program, this isn't simple.
>
> Thank you,
> Gila.
>