Use of faldo:Region in for mapping chromosomal band positions

[Matthew Brush]

My more general question here is when can a feature being described be linked directly to a faldo:Postion  - as opposed to first being linked to a faldo:Region using the faldo:location property, and then this faldo:Region getting linked to a faldo:Position? I understand the reasoning behind using faldo:Regions to separate features from their positions on reference sequences - but wondering if cases where the feature is a defined part of some reference genome allow it to be considered as a faldo:Region, and linked directly to its position.

My specific use case is translating the UCSC Cytogenic Band dataset to rdf. The dataset is here. In this dataset, the features being described are cytogenic bands in the hg19 human genome build. The positions of these bands are defined relative to the chromosomes in this hg19 build. Because these bands are part of a specific reference build, can I consider these bands to themselves be faldo:Regions, and link the band directly to faldo:positions. For example, for the band 10p11.22, could I say:

ucsc:hg19chr10p11.22  a  so:chromosome band,   faldo:Region ;
    rdfs:label "chr10p11.22 (hg19)" ;
    faldo:begin  _:PositionX
    faldo:end   _:PositionY
 

Or is it still best to link the bands to faldo:Regions, and then link the faldo:Region to faldo:Positions?
For example:

ucsc:hg19chr10p11.22  a  so:chromosome band ;
    rdfs:label "chr10p11.22 (hg19)" ;
    faldo:location  _:Region1 .

_:Region1  a  faldo:Region ;
    faldo:begin  _:PositionX
    faldo:end   _:PositionY