[Biojava-l] Re: CORBADEV: LSR news - new RFP about biomolecular entities

[Philip Lijnzaad]

Matthew> Afternoon.
>> I have just two suggestions for addition:
>> 
>> - a type for representing floating point information on sequence positions,
>> such as hydrophobicity plots. I am not going to insist on this, because the
>> (nearly) obvious solution is to just use a IDL::float. Or was that double ...
>> 

Matthew> This can be cleanly represented as an alignment containing a
Matthew> sequence (protein) and hydrophobicity (floats). This would require
Matthew> either a special type within alignments that has numbers, or a
Matthew> re-working of how sequences are represented so that an individual
Matthew> symbol could be a float (the neater solution). This may be covered
Matthew> by the BioSequence alphabet extensions. I don't know how the
Matthew> alignment object works in BSA so I may be way of mark.

Hi Matthew,

thanks for your feedback. My take on this is that in the simple case of a
hydrophobicity plot this would be unnatural, because you don't actually align
the plot to the sequence with scores and gaps and all; you just compute it
and add it as an annotation. 

For doing more complex things like secondary structure or transmembrane
predictions, I would still not use Alignment, but using just one float (or
double) per position is probably not going to be enough. Typically there will
be a confidence value there as well. So I now wonder where to draw the line:
if a standard wants to standardize things for float values, should it then
not also standardize things for other values? 

(But anyway, this was just to see if a type is needed for this, and if so,
put it in the requirements for the standard. The submitters of proposals will
then have to see how best to represent this).

Cheers,

                                                                      Philip

-- 
When C++ is your hammer, everything looks like a thumb. (Steven Haflich)
-----------------------------------------------------------------------------
Philip Lijnzaad, lijnzaad@ebi.ac.uk \ European Bioinformatics Institute,rm A2-24
+44 (0)1223 49 4639                 / Wellcome Trust Genome Campus, Hinxton
+44 (0)1223 49 4468 (fax)           \ Cambridgeshire CB10 1SD,  GREAT BRITAIN
PGP fingerprint: E1 03 BF 80 94 61 B6 FC  50 3D 1F 64 40 75 FB 53