Skip to content


huaiyumi edited this page Jun 11, 2016 · 11 revisions

AF phenotype glyph survey

In SBGN Activity Flow, phenotype and perturbation are both activity nodes, and are represented by glyphs identical to those of perturbing agent and phenotype in PD, respectively. The syntax of both glyphs is identical to that in PD also, i.e., perturbation can only be the source of an influence arc, and phenotype can only be the target.

There are two issues related to this.

  1. In PD, perturbing agent is an EPN, while phenotype is a process node, while in AF, both perturbation and phenotype are activity nodes (processes). Using the same glyph for two semantically different concept is confusing.

  2. In AF, sometime a phenotype can be a source of the different arc, for example, Na channel activity -> membrane depolarization -> K channel activity -> membrane repolarization. The current AF syntax makes it difficult to represent such pathways.

The topic was discussed at the HARMONY in April, and a few solutions have been suggested, which are summarized below.

A. Solution for the glyph of perturbation

Remove perturbation as an activity node. Instead, use it as a decoration of an activity node like macromolecule or nucleic acid feature.

Examples: the glyph, PPAR pathway.

B. Solutions of phenotype

Below is a summary of the suggestions gathered at the HARMONY

  1. Remove phenotype glyph, and use activity node to represent it.
    • Example
    • Problem: It loses the specificity to represent something that is observed.
  2. Remove phenotype glyph, use activity node with a specific decoration that reflects the fact that it can be observed.
    • Example
    • Problem: a new glyph has to be designed, and its meaning is similar to that of phenotype.
  3. Remove phenotype glyph, use activity node with a phenotype glyph as a decoration.
    • Example
    • Problem: phenotype glyph is a process node in PD, and semantically it should not be used as a decoration.
  4. Keep phenotype glyph, and it can only be the output of an influence arc, but design a new influence arc such as “causation”.
    • Example
    • Problem: need to create a new arc glyph and its meaning could potentially be the same as existing ones.
  5. Allow phenotype glyph to be both input and output of influence arcs.
    • Example
    • Problem: It is not consistent with that of PD. In PD, a phenotype can only be the boundary of the pathway, i.e., the target of an arc.
  6. Keep the current spec, i.e., phenotype glyph can only be the output of an arc.

Results of the voting

1 vote - proposal 1

1 vote - proposal 2

0 vote - proposal 3

0 vote - proposal 4

3 votes - proposal 5

2 votes - proposal 6

No decision was made. Need further discussion.

Clone this wiki locally
You can’t perform that action at this time.