Linking RDF to Cheminformatics and Proteochemometrics
Author Information
Author(s): Egon Willighagen, Jonathan Alvarsson, Annsofie Andersson, Martin Eklund, Samuel Lampa, Maris Lapins, Ola Spjuth, Jarl ES Wikberg
Primary Institution: Uppsala University, Department of Pharmaceutical Biosciences
Hypothesis
Can RDF technologies be effectively integrated into molecular chemometrics methods?
Conclusion
Existing RDF standards can be integrated into molecular chemometrics methods, enhancing data aggregation and analysis.
Supporting Evidence
- RDF technologies can link non-numerical molecular knowledge representation to numerical representations used in statistical modeling.
- Bioclipse was used to integrate various RDF functions and visualize chemical data.
- Statistical modeling showed improved predictive performance when assay confidence was considered.
Takeaway
This study shows how a special data format called RDF can help scientists better understand and analyze chemical information.
Methodology
The study utilized RDF technologies, SPARQL for data retrieval, and statistical modeling methods including Bayesian weighted ridge regression.
Limitations
The RDF standards do not specify how chemical information should be modeled, leading to potential inconsistencies.
Statistical Information
Confidence Interval
(7.6; 12.8)
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website