The molecular entities in linked data dataset

Area of Science:

• Cheminformatics
• Bioinformatics
• Linked Data

Background:

• Chemical entities are fundamental to molecular research.
• Existing datasets may lack rich semantic relationships or standardized formats.
• Need for accessible, structured chemical data for computational analysis.

Purpose of the Study:

• Introduce the Molecular Entities in Linked Data (MEiLD) dataset.
• Provide a comprehensive, semantically rich dataset of chemical entities.
• Facilitate research and development in cheminformatics and bioinformatics.

Main Methods:

• Data generation using SDFEater tool.
• Utilizing Resource Description Framework (RDF) data model.
• Enrichment with ontologies like Chemical Vocabulary for Molecular Entities (CVME) and SKOS.
• Data provided in JSON-LD format.

Main Results:

• Creation of the MEiLD dataset containing 349,960 small chemical entities.
• Dataset preserves hierarchical and associative semantic relations.
• Publication of the MEiLD dataset, SHACL schema, and CVME ontology.
• Data availability via Mendeley Data.

Conclusions:

• The MEiLD dataset provides a valuable, semantically rich resource for computational chemistry and biology.
• Standardized linked data format enhances data integration and analysis.
• Accessibility of the dataset and associated schemas promotes wider adoption and research.