dcGOR: an R package for analysing ontologies and protein domain annotations

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• The dcGO database provides protein domain annotations using various ontologies, including Gene Ontology.
• Existing bioinformatics tools lack specific support for domain ontology analysis, limiting the potential of resources like dcGO.
• There is a need for integrated statistical and graphical support for analyzing protein domain annotations within biological ontologies.

Purpose of the Study:

• To introduce 'dcGOR', an open-source R package designed for analyzing ontologies and protein domain annotations.
• To provide a flexible and functional bioinformatics tool for the dcGO database and similar resources.
• To enable in-depth analysis of functional, phenotypic, and disease relevance of protein domains and their associated ontologies.

Main Methods:

• Implementation of a novel data structure within the R environment to represent domains, ontologies, and annotations.
• Development of mining facilities for each ontology, including domain-based enrichment analysis, semantic similarity network construction, and statistical significance analysis.
• Support for high-performance parallel computing to optimize analysis runtime.

Main Results:

• The dcGOR package facilitates domain-based enrichment analysis and visualization.
• It enables the construction of domain (semantic similarity) networks and statistical significance networks.
• The package supports the analysis of user-provided ontologies and annotations from various sources (SCOP, Pfam, InterPro, Rfam).

Conclusions:

• dcGOR provides essential statistical and graphical support for domain ontology analysis, enhancing the utility of resources like dcGO.
• The package offers flexibility, allowing users to analyze their own ontologies and annotations, extending its applicability.
• dcGOR is freely available on CRAN and GitHub, promoting its adoption and use within the bioinformatics community.