Semantic Web integration of Cheminformatics resources with the SADI framework

Area of Science:

• Computational Chemistry
• Bioinformatics
• Cheminformatics

Background:

• Chemical research exhibits poor interoperability of computational resources and databases due to diverse, independent efforts.
• Existing solutions partially address database interoperability but fail to integrate computational resources effectively.
• This hinders reproducibility and necessitates expert-driven, application-specific workflows, especially in interdisciplinary fields like systems chemistry.

Purpose of the Study:

• To leverage Semantic Web Services (SWS) for integrating diverse chemical computational and database resources into a unified, machine-understandable system.
• To address the limitations in computational resource interoperability and enhance the reproducibility of computational experiments in chemistry.

Main Methods:

• Developed a prototype framework of Semantic Automated Discovery and Integration (SADI) using SWS.
• Exposed the Chemistry Development Kit's Quantitative Structure-Activity Relationship (QSAR) descriptor functionality as SADI SWS.
• Utilized formal ontology-defined input/output classes and RDF graphs for service reasoning and task completion via SPARQL queries.

Main Results:

• Demonstrated automatic reasoning about services and reference information for computational tasks.
• Successfully performed QSAR analysis to predict drug-likeness using a formal ontology.
• Showcased parameter-based control over SADI SWS execution and demonstrated value through service reuse and integration.

Conclusions:

• The SADI framework using SWS can revolutionize the distribution of computational resources in chemistry.
• Enveloping chemical computational resources as SADI SWS facilitates interdisciplinary research.
• Enables defining computational problems via ontologies and logic, replacing cumbersome, application-specific workflows.