Neil Chalk wrote:
>> Dear NG,
>> I am doing a Machine Learning project which aims to predict splice sites
> of proteins. For the requirements analysis side of things I need to now
> what would be useful to researchers in this field and what systems are
> currently being used (and how well they fill the need). I have had a
> look at the software lists on sites such as The Bioinformatics Resource
> (http://www.hgmp.mrc.ac.uk/CCP11/) .. but I am not a specialist so I
> have probably missed some relevant packages.
Neil,
Do you really mean protein splice sites (ie Inteins)?? Or are you
looking at DNA splice sites which lead to the mRNA which gets translated
into proteins?
If you're interested in Inteins, then have a look at;
http://www.uk.neb.com/neb/inteins/int_bibliog.html
I'm not aware of any programs which will predict intein splice sites.
If you're actually looking at DNA splicing, then there are many tools
available for this purpose. There are specific splice donor and
acceptor sequences which can be searched for. Also, much information
can be gained by making alignments of genomic DNA and cDNA (made from
mRNA) sequences. The tool I find most useful for this purpose is NIX at
HGMP. This runs several predictions and searches, and gives you back an
interactive summary view.
NIX can be found at;
http://www.hgmp.mrc.ac.uk/Registered/Webapp/nix/
You need an HGMP account to use NIX - but this is free to academics.
Hope some of this is useful.
TTFN
Simon.