In SBOL, when should I annotate a DNA sequence vs. making sub-components?

Question

When representing a DNA sequence in SBOL as a ComponentDefinition, you can mark things like promoters and coding sequences in two different ways:

• either as a SequenceAnnotation in the ComponentDefinition, or
• as a Component linking to a ComponentDefinition for the sub-comoponent

When should you choose one over the other?

Answer 1

Typically you want to use a ComponentDefinition when an sub-structure is something you might want to pull out and re-use in another genetic design, and a SequenceAnnotation whenever it is not.

For example:

• Promoters, terminators, and protein coding sequences are typically best represented with their own ComponentDefinition, since these are often pulled out and re-combined in new genetic designs. Representing such an element with a ComponentDefinition makes it easy to find the commonalities between designs.
• Assembly scars, binding sites, and cut sites are typically best represented with a SequenceAnnotation because they are typically not very interesting except in the context of their surrounding genetic context. Representing such an element with a SequenceAnnotation helps keep them associated tightly with that context.

Not every case is cut and dried, however. For example, ribosome entry sites are often treated as separable components, but their performance is very tightly tied to the specifics of the coding sequence that they modulate. Thus, in some cases it may make more sense to represent them separately as a ComponentDefinition and in others to represent them as a SequenceAnnotation on a joint RES/CDS ComponentDefinition.