Semanticscience Integrated Ontology (SIO)

The Semanticscience Integrated Ontology (SIO) is a simple yet comprehensive upper-level ontology that provides foundational types and relations for consistent semantic knowledge representation across physical entities, processes, and information constructs. SIO defines core abstract concepts such as objects, processes, attributes, and information entities, enabling researchers to build domain-specific ontologies with consistent semantic foundations. The ontology supports interconnection of diverse knowledge domains through its generalized entity and relationship types, facilitating interoperability across heterogeneous biological and biomedical ontologies. SIO serves as the semantic backbone for major linked data projects including Bio2RDF (which integrates biological databases) and SADI (Semantic Automated Discovery and Integration framework). The ontology enables semantic reasoning and automated knowledge discovery by providing explicit typing and relationship hierarchies that computationally systems can interpret.

Example Usage: Use SIO classes to type entities in a knowledge graph (e.g., SIO:Protein, SIO:Gene, SIO:Organism) and relate them via SIO properties (e.g., SIO:is-derived-from, SIO:encodes, SIO:has-function) for automated biomedical data integration and discovery.

Metrics & Statistics

Graph Statistics

Total Nodes

7811

Total Edges

15701

Root Nodes

18

Leaf Nodes

4921

Knowledge Coverage Statistics

Classes

1726

Individuals

0

Properties

212

Hierarchical Metrics

Maximum Depth

20

Minimum Depth

0

Average Depth

6.67

Depth Variance

12.94

Breadth Metrics

Maximum Breadth

186

Minimum Breadth

1

Average Breadth

63.71

Breadth Variance

3373.16

LLMs4OL Dataset Statistics

Term Types

0

Taxonomic Relations

2019

Non-taxonomic Relations

65

Average Terms per Type

0.00

Usage Example

Use the following code to import this ontology programmatically:

from ontolearner.ontology import SIO

ontology = SIO()
ontology.load("path/to/SIO-ontology.owl")

# Extract datasets
data = ontology.extract()

# Access specific relations
term_types = data.term_typings
taxonomic_relations = data.type_taxonomies
non_taxonomic_relations = data.type_non_taxonomic_relations