[Bioperl-l] GeneStructure & Promoter(s)

[Elia Stupka]

After all the previous e-mails, I would vouch for:

RegulatoryElements (RE): container objects for regulatory SeqFeatures

Promoter (P): SeqFeature contained in the above, but also contains a set
of subSeqFeatures (such as TATA box, TF binding sites,etc.)

Enhancer (E): SeqFeature contained in the above

As an extra feature compared to standard SeqFeatures, these objects should
allow for interactions between them, such as interactions between TF
binding sites, or enhancers and promoters. That could be defined as simple
plus, minus or zero, or as a continuous value from -1 to +1. (Then again I
guess this would show the need for a transcription factor object).


	RE
	|
       / \
      E   P
         /|\
        / | \
       /  |  \
    TATA TF1  TF2
          |    |
          ------


I think the fact that "nobody knows what is going on exactly" should not
be a reason to discourage us. On the contrary, for the time being, we
should make a structure flexible enough for people to be able to use these
objects to describe hypotheses, not only "facts". At the end of the day
the code in bioperl compared to the one in EnsEBML I see more as a
"toolkit" for biologists than for large genomics firms/centres.

Elia

**************************
tel:    +44 1223 49 44 31
mobile: +44 7971 59 03 69
fax:    +44 1223 49 44 68
**************************