Stress Knowledge Map

Supported by the ADAPT project

PSS database schema

Naming conventions

Name format
  • Node labels: UpperCamelCase
  • Relationship types: UPPER_SNAKE_CASE
  • Property keys: snake_case
Edge
Relationship between two nodes
Source
source (start) node of a relationship.
Target
target (end) node of a relationship.

Node labels

A node can have multiple labels. In general, the primary labels have a constraint of a unique name within the label.

Primary Labels

Primary label Secondary labels (*=optional) Description
Complex Node Complexes that can contain plant, foreign and/or metabolite components
Metabolite Node, MetaboliteFamily* Chemical entities
Family Node, [PlantAbstract | PlantCoding | PlantNonCoding] Gene family (set of all genes within the family)
Clade Node, [PlantAbstract | PlantCoding | PlantNonCoding] Set of genes defined by sequence similarity
FunctionalCluster Node, [PlantAbstract | PlantCoding | PlantNonCoding] Set of genes with the same function
Process Node
ForeignEntity Node, Foreign Biological foreign entity, such as a pathogen species
ForeignCoding Node, Foreign Protein coding agent of an external factor, such as a pathogen gene
ForeignNonCoding Node, Foreign Non-coding agent of an external factor, such as a pathogen gene
ForeignAbstract Node, Foreign Undefined or unspecified agent of an external factor, such as a pathogen gene
ForeignAbiotic Node, Foreign Abiotic factors (e.g. heat)
Reaction Node Reaction node defines a reaction or interaction between entities
Condition Condition that affects reaction(s)

Secondary Labels

Label Description
Node All nodes
MetaboliteFamily A group of similar chemical entities (e.g. reactive oxygen species)
PlantAbstract Undefined or unspecified plant gene (e.g. without standard gene identifier)
PlantCoding Plant gene with protein gene product
PlantNonCoding Plant gene without protein gene product (e.g miRNA)
Plant All plant specific nodes (PlantAbstract, PlantCoding, PlantNonCoding)
Foreign External factors that are neither plant nor metabolite related

Edge types

Each edge only has one type.

Hierarchical related edges

Type Description Edge example
COMPONENT_OF Source node is a component of target node (Complex) (JAZ) ―[:COMPONENT_OF]→ (JAZ|DELLA)
TYPE_OF Source node (Metabolite) is a type of target node (MetaboliteFamily) (O3) ―[:TYPE_OF]→ (ROS)
AGENT_OF Source node (Foreign[*]) is an agent of target node (ForeignEntity) (VPg) ―[:AGENT_OF]→ (potyvirus)
HAS_CLADE Source node (Family) has child clade target node (Clade) (AGO) ―[:HAS_CLADE]→ (AHP1,2,3,4,5)
TAKES_PART Source node (Family) takes part in reactions via target node (FunctionalCluster) (ACS) ―[:TAKES_PART]→ (ACS6)
CONDITION_INPUT Source node form part of the condition target node (Condition) defines (JAZ) ―[:CONDITION_INPUT]→ (NPR1 high & JAZ high)

Reaction related edges

Type Description Edge example
ACTIVATES Source node activates target node (Reaction) (CYP94) ―[:ACTIVATES]→ (rx00042)
INHIBITS Source node inhibits target node (Reaction) (miR6022) ―[:INHIBITS]→ (rx00308)
SUBSTRATE Source node is a substrate in the reaction, and is consumed/catalysed by reaction (JA-Ile) ―[:SUBSTRATE]→ (rx00042)
PRODUCT Target node is a product in the reaction, and is produced/resulted from reaction (rx00042) ―[:PRODUCT]→ (12-OH-JA-Ile)
TRANSLOCATE_FROM Source node is translocated from source_compartment by reaction (sister to SUBSTRATE) (SA) ―[:TRANSLOCATE_FROM]→ (rx00072)
TRANSLOCATE_TO Target node is translocated to target_compartment by reaction (sister to PRODUCT) (rx00072) ―[:TRANSLOCATE_TO]→ (SA)

Properties

'-' indicates all nodes and all edges
Property name Applies to (labels/edge types) Type Note
added_by - string
creation_date - string
name - string
components Complex list
classification ForeignEntity string Phylogenetic classification
species ForeignEntity string
family FunctionalCluster string
sequence FunctionalCluster string Only in case of gene not existing in genome model (of GoMapMan)
short_name FunctionalCluster string
additional_information Node string
curated Node bool
external_links Node list unknown, invented:<reason>, or <database>:<database identifier>, where database includes doi, pmid, aracyc, chebi, pubchem, kegg, etc
model_version Node string
pathway Node string One of defined pathways
model_status Node excluding Reaction string
description Node, Condition string
<species>_homologues Plant list list of species specific homologues
species Plant list Subset of defined species
evidence_sentence Reaction string
experimental_techniques Reaction string
reaction_effect Reaction string activation or inhibition
reaction_mechanism Reaction string
species Reaction list Subset of defined species, and "all"
trust_level Reaction string R1;R2;Rx...
reaction_id Reaction, all reaction related edges string
reaction_type Reaction, all reaction related edges string One of defined reaction types
source_form ACTIVATES, INHIBITS, SUBSTRATE, TRANSLOCATE_FROM string One of defined node forms
source_location ACTIVATES, INHIBITS, SUBSTRATE, TRANSLOCATE_FROM string One of defined locations
source_organ ACTIVATES, INHIBITS, SUBSTRATE, TRANSLOCATE_FROM string One of defined plant organs (leaf, stem, root)
target_form PRODUCT, TRANSLOCATE_TO string One of defined node forms
target_location PRODUCT, TRANSLOCATE_TO string One of defined locations
target_organ PRODUCT, TRANSLOCATE_TO string One of defined plant organs (leaf, stem, root)

Incomplete properties: gomapman annotations.

Indices

Indices are generated across nodes with the same label to allow faster searching.

In progress

Plant node hierarchy

Families contain all genes associated with the family, across plant species.

Clades are subsets of families, as defined by sequence similarity.

FunctionalCllusters are subsets of (possibly multiple) families, as defined by participation in reactions (i.e. their function).

Of the three layers, only FunctionalClusters can participate in reactions.

Plans - incorporate Clades as "parents" of FunctionalClusters, instead of families

Defined reaction types

The database schema is based on a representations of chemical reactions. In a simple example, the following reaction A + B → C (catalysed by e):

can be represented in the graph database (using only nodes and edges) as follows:

Edges (arrows) and reaction nodes are reaction specific. Nodes other than reactions can take part in multiple reactions.

Using this schematic, the database can contain the following reaction types:

binding/oligomerisation
dissociation
catalysis
degradation/secretion
protein deactivation
protein activation
transcriptional/translational activation
transcriptional/translational repression
translocation
unknown

Please note that NEWT is a client-side application, and therefore relies on your local computer capabilities. Given the relatively large size of PSS, rendering the network may take a while.