01 / DATA MODEL
↗
RDF
Resource Description Framework
Represents statements as subject–predicate–object triples. These statements form a directed, labeled graph.
ONTOLOGY / SEMANTIC TECHNOLOGIES
Semantic
Technologies.
KNOWLEDGE SYSTEMS / 001
Explore the technologies that enable information to carry explicit meaning, allowing data from different sources to be connected, queried and interpreted using shared standards.
THE SEMANTIC WEB / CORE PRINCIPLE
Data becomes more useful when relationships have meaning.
Semantic technologies provide methods for representing information through identifiable entities, explicit relationships and formally defined concepts. They help systems exchange information without relying solely on identical column names or isolated keywords.
ENTITY / RELATIONSHIP / MEANING
FIG 01 / RDF GRAPH
What are semantic technologies?
Semantic technologies are approaches to structuring, representing, linking and processing information in ways that make the intended meaning more explicit. They commonly use shared identifiers, vocabularies, ontologies and graph-based representations.
Traditional databases store information efficiently, but information from different organizations can be difficult to combine when terms, identifiers and schemas differ. A biological laboratory, for example, might describe a gene using a local identifier while another dataset uses a standardized external accession number.
Semantic technologies can address this challenge by connecting entities to persistent identifiers and describing relationships using agreed vocabularies. Such representations make it easier to combine data across compatible systems and recover information through graph-based queries.
The goal is not to make computers understand information exactly as humans do. Rather, it is to provide sufficiently explicit structure and formal meaning for reliable computational processing.
003 / THE TECHNOLOGY STACK
The languages behind structured knowledge.
Different W3C standards provide complementary capabilities for describing, organizing, querying and interpreting linked information.
02 / SCHEMA
↗
RDFS
RDF Schema
Defines basic vocabulary for classes, properties, subclass hierarchies, domains and ranges.
03 / ONTOLOGY
↗
OWL
Web Ontology Language
Adds formal constructs for modeling classes, properties and logical restrictions that support automated reasoning.
04 / QUERY
↗
SPARQL
Graph Query Language
Retrieves and transforms RDF data by matching graph patterns, with support for filtering, aggregation and more.
05 / VOCABULARY
↗
SKOS
Simple Knowledge Organization System
Supports structured thesauri, concept schemes, labels, broader and narrower relationships, and vocabulary mappings.
06 / PUBLISHING
↗
LD
Linked Data
Uses web identifiers and links to publish data that can be connected and reused across independent sources.
004 / FROM CONCEPT TO DATA
Understanding the RDF triple.
RDF represents information in statements containing three components: a subject, a predicate and an object. The subject identifies the resource being described. The predicate specifies a relationship, while the object identifies a related resource or value.
For example, a dataset can describe a particular gene as being associated with a named biological process. If the same identifiers and relationships are reused across sources, their information can be linked together.
The simplified Turtle example illustrates how machine-readable statements are written. The identifiers are hypothetical examples, not references to real experimental data.
RDF / TURTLE EXAMPLE
MODEL 001
@prefix ex: <https://example.org/bio/> . @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> . @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> . ex:Gene a rdfs:Class . ex:BiologicalProcess a rdfs:Class . ex:geneA a ex:Gene ; rdfs:label "Example gene A" ; ex:associatedWith ex:processB . ex:processB a ex:BiologicalProcess ; rdfs:label "Example process B" .
Illustration only. A production
ontology should formally
define the relationship
ex:associatedWith, reuse
relevant established
identifiers, and document
the supporting evidence.
RDF is a graph data model. Turtle is one textual serialization of RDF; RDF/XML, JSON-LD and other compatible serializations are also used.
005 / SCIENTIFIC COMPARISON
Which technology does what?
These technologies work together, but they are not interchangeable. Understanding their roles helps researchers choose the right representation and analysis tools.
| Technology | Main Function | Scientific Example |
|---|---|---|
| RDF | Represents graph statements | Connect a gene identifier to an annotation |
| RDFS | Defines basic class and property semantics | State that Neuron is a subclass of Cell |
| OWL | Expresses richer logical ontology axioms | Define class relationships and restrictions |
| SPARQL | Queries RDF graphs | Find genes linked to a selected biological process |
| SKOS | Organizes concept schemes and controlled vocabularies | Maintain a multilingual scientific terminology system |
| Linked Data | Connects identified resources across datasets | Link molecular records with biomedical databases |
006 / KNOWLEDGE PIPELINE
From isolated records to connected data.
Semantic integration is a process that starts with identifying the concepts shared by different sources and ends with structured information that can be queried and interpreted.
01
Identify the domain concepts
Determine which entities, properties and relationships need to be represented, such as genes, proteins, samples or biological processes.
02
Choose shared identifiers and vocabularies
Reuse established terminology and identifiers wherever appropriate to reduce ambiguity and support interoperability.
03
Represent the relationships
Transform suitable source information into RDF statements or compatible graph representations, preserving relevant provenance and context.
04
Validate and query the data
Check the structure and expected constraints. Query linked records using SPARQL and, where appropriate, apply ontology reasoning.
05
Interpret the scientific results
Connect retrieved relationships to their biological evidence. A graph connection can support a hypothesis, but does not automatically establish biological causality.
007 / APPLICATIONS IN LIFE SCIENCES
Semantic technologies in biomedical research.
Biomedical research integrates information from experiments, molecular databases, publications and clinical terminology systems. Semantic technologies help describe and interconnect this heterogeneous information.
In biotechnology, such integration can support data discovery, annotation, cross-database comparison and the systematic interpretation of scientific knowledge.
Gene and pathway integration
Connect gene identifiers, protein annotations and biological processes using standardized terms. Such relationships can support functional enrichment studies and biological data interpretation.
Phenotype and disease knowledge
Integrate disease-associated information with structured phenotype descriptions. Biomedical ontologies help researchers compare findings across datasets using shared terminology.
Research interoperability
Represent experimental metadata and scientific entities consistently across databases, allowing software tools to interpret compatible records using explicit identifiers and relationships.
Knowledge-aware AI systems
Knowledge graphs and ontologies can provide structured context for search, information retrieval and some hybrid reasoning systems. The reliability of resulting AI outputs still depends on the underlying data and validation.
008 / TECHNICAL REFERENCE LIBRARY
Explore the official standards.
Primary specifications and reference documentation from the World Wide Web Consortium provide authoritative descriptions of these technologies.
01 /
W3C ↗
RDF 1.1 Primer
Introduction to RDF triples, identifiers, graph data and basic representation.
04 /
W3C ↗
SPARQL 1.1 Overview
Querying and updating RDF graph data using standardized query facilities.
05 /
W3C ↗
SKOS Reference
Standard vocabulary for knowledge organization systems and concept schemes.
009 / FREQUENT QUESTIONS
Semantic technologies, explained.
What is the difference between RDF and OWL?
RDF provides a graph-based data model for expressing statements. OWL provides additional formal ontology constructs and logical semantics that support more expressive knowledge representation and reasoning.
Is SPARQL a database?
No. SPARQL is a query language and protocol family for RDF data. Graph stores and other systems may implement SPARQL query capabilities.
Is a knowledge graph the same as an ontology?
Not necessarily. A knowledge graph represents entities and their connections. An ontology describes concepts and relationship semantics. A knowledge graph may use an ontology to make its structure more consistent and interpretable.
Can semantic technologies improve biomedical research?
They can help researchers integrate heterogeneous data, standardize terminology and retrieve linked scientific information. Their effectiveness depends on data quality, appropriate modeling and accurate domain knowledge.
Do semantic technologies automatically prove scientific relationships?
No. Formal reasoning can derive logical consequences from defined axioms, but it does not independently establish experimental truth. Scientific claims require appropriate evidence and validation.