Representing Exon boundaries

Jalview Discussion
1

Hi,

Is there any formal way to represent exon boundaries in a protein alignment in Jalview?

Cheers,

Albert.

2

Hello Albert.

Albert Vilella wrote:

Is there any formal way to represent exon boundaries in a protein
alignment in Jalview?

Actually, no. There is support for this kind of annotation internally,
but it is currently only used for the linked highlighting of DNA and
protein views, and mapping features and annotation.

I have also noticed cases in the literature where exon boundaries have
been added to an alignment as vertical bars, and at the time realised
that it would be useful to do this. I guess it would be possible to mock
up the annotation by creating a group on the sequence for each exon
region on the peptide product - but this is pretty painful to do by
hand, and would be alignment annotation - rather than sequence
annotation, so would be broken as soon as the alignment was edited.

This will be a good addition to the jalview feature list - no guarantees
about when it is done .. of course.. but perhaps sometime soon (unless
there are any volunteers!).

thanks.
Jim

···

--
-------------------------------------------------------------------
J. B. Procter (ENFIN/VAMSAS) Barton Bioinformatics Research Group
Phone/Fax:+44(0)1382 388734/345764 http://www.compbio.dundee.ac.uk
The University of Dundee is a Scottish Registered Charity, No. SC015096.

3

Hi,
I am glad this topic comes up because it is something I have done more or less by hand in a couple of cases, and it is inconvenient.

However, I just want to point out that exon boundaries can occur within codons, so it is not straightforward to represent it on the amino acid level. In fact, it is not obvious how to visualize it.

    Lars,
    a happy Jalview user

James Procter wrote:

···

Hello Albert.

Albert Vilella wrote:
  

Is there any formal way to represent exon boundaries in a protein
alignment in Jalview?
    

Actually, no. There is support for this kind of annotation internally,
but it is currently only used for the linked highlighting of DNA and
protein views, and mapping features and annotation.

I have also noticed cases in the literature where exon boundaries have
been added to an alignment as vertical bars, and at the time realised
that it would be useful to do this. I guess it would be possible to mock
up the annotation by creating a group on the sequence for each exon
region on the peptide product - but this is pretty painful to do by
hand, and would be alignment annotation - rather than sequence
annotation, so would be broken as soon as the alignment was edited.

This will be a good addition to the jalview feature list - no guarantees
about when it is done .. of course.. but perhaps sometime soon (unless
there are any volunteers!).

thanks.
Jim

--
Stockholm Bioinformatics Center
and School of Computer Science and Communication
http://www.csc.kth.se/~arve

4

Hi Lars..

Lars Arvestad wrote:

I am glad this topic comes up because it is something I have done more
or less by hand in a couple of cases, and it is inconvenient.

However, I just want to point out that exon boundaries can occur within
codons, so it is not straightforward to represent it on the amino acid
level. In fact, it is not obvious how to visualize it.

Very good point. I suspect there are one or two different conventions,
in fact. With regards to handling the underlying data, within the VAMSAS
datamodel we explicitly built in the concept of annotation being either
at a discrete position (i.e. base or residue) in a coordinate system, or
after it, in order to properly reflect inclusive or exclusive
boundaries. I think it might be possible to utilise the same trick for
proteins. Codon boundaries are at a residue, or after it. Anything more
would have to be a subtype. Of course, the original issue stands -
jalview won't render exon boundaries after a residue. It could,
however, be 'hacked' in as a special feature representation:
'disulphide' feature type rendering is done this way.. ( as a hack - but
in principle, the rendering model could re-implemented as something more
generic that is driven by some ontology/look up table of feature types
and styles). One could almost imagine positioning a bar either mid-left,
mid-right, or at the right of the residue symbol.

j.

···

--
-------------------------------------------------------------------
J. B. Procter (ENFIN/VAMSAS) Barton Bioinformatics Research Group
Phone/Fax:+44(0)1382 388734/345764 http://www.compbio.dundee.ac.uk
The University of Dundee is a Scottish Registered Charity, No. SC015096.