Pathways

A pathway model is a dynamic network of entities (for example, proteins, genes, drugs, and so on) connected by influence relationships. These interaction relationships describe the activation or inhibition effects that entities can have on each other. Ultimately, simulations help to explain the impact of cellular stimuli on a cellular phenotype or to computationally assess research hypotheses.

Siena. S et al., J Natl Cancer Inst. 2009 Oct 7;101(19):1308-24. doi: 10.1093/jnci/djp280. Epub 2009 Sep 8.

You can merge two pathway models to create a new model that retains the unique entities and interactions from the input models. If the entities in the input models have more than one identifier in total, you can choose to rename the resulting merged entity and change it to a complex entity type. For example, if both the input models share a Uniprot identifier but one of them also has another identifier, you might want the name to reflect their combination. If unique entities from the input pathway models have different annotations or evidence information, the merge automatically combines these into the new model. The resulting merged pathway model includes recalculated formulas according to the relationships in the model. The positions of entities and interactions in the merged pathway model use the average positions of the pair of input merged pathway models.

An annotation is a human readable piece of text. This provides complementary information supporting or documenting the role of an entity or an interaction. It can also provide additional details on the context, or to qualify a decision (for example, share contradictory arguments on a modeling choice). The annotation categories are:

Cross-reference annotations describe reference entries for the entity in external public domain resources.
Disease annotations document the diseases the entity is associated with.
Hallmark annotations document characteristic disease phenotypes or biological processes associated with the entity.
Genetic alteration annotations specify types of alterations associated with the entity.
Other annotations can specify additional information related to this entity.

Disease, hallmark, and genetic alteration annotation categories have predefined candidate values, but you can add your own values. You can use evidence links to support all annotations categories other than cross-references. The annotations automatically interpret links to any of the reference databases and display them in the format source : id. Annotations do not interpret links to other reference resources, displaying them using the raw URL.

Evidence is a link or a reference to a document supporting an annotation. (for example, a URL to a pubmed article). When you paste a URL into the evidence field, this automatically generates links to some reference databases.


Reference Database	Referenced Concepts
interpro	Protein domains
uniprot	Proteins
obo.go	Gene ontologies
kegg.compound	Compounds
kegg	Pathways
biomodels	Pathways
pubmed	Scientific articles
pubmedcentral	Scientific articles
taxonomy	Uniprot taxonomy
hgnc	Gene nomenclature
reactome	Pathways
intact	Protein-protein interactions
pdb	Protein crystallographic structures
doi	Digital object identifier
omim	Diseases

The formula is the expression of an entity's behavior. This determines the entity's state at a given step of a simulation. The formula is a Boolean function reflecting the influence of upstream entities. See Simulations.

The behavior type describes which of the two types of formula applies to an entity:

Automatic Formula: Describes the behavior of an entity solely based on the topology of incoming interactions (the AND combination of inhibiting and activating influences on this entity).
Custom Formula: It can be edited by the user to modify the entity behavior and to override the automatic formula. Changes made in the custom formula do not modify the topology and are not propagated to the interactions and interactions effects.

The initial state reflects the state of an entity before performing any simulations. You can modify this to:

Tune the model for a specific context (for example, a disease, a tissue).
Reflect a molecular perturbation during the design of an experiment (for example, a gain of function mutation).