DOSE is an R/Bioconductor package for semantic and enrichment analysis based on the Disease Ontology (DO). It enables biologists to explore disease similarities and gene function similarities in a disease context. DOSE supports semantic similarity computations between DO terms and genes, and enrichment analyses including hypergeometric model and gene set enrichment analysis (GSEA) to discover disease associations from high-throughput data. It also provides visualization functions for similarity and enrichment analyses, which are not available elsewhere. DOSE allows users to measure semantic similarity among DO terms and genes using several information-content and graph-structure based algorithms. It supports hypergeometric test and GSEA to evaluate disease relevance of biological experiments. DOSE provides several DO-specific visualization functions to produce highly customizable, publication-quality figures of similarity and enrichment analyses. The package includes functions for comparing disease associations among different gene sets. DOSE was developed using the R statistical computing language and is released within Bioconductor. It provides five algorithms for DO and gene semantic similarity measurements, hypergeometric test for identifying significant disease association of gene list, GSEA for interpreting genome-wide expression profiles in disease context, and comparison of significant disease associations among different gene sets. DOSE facilitates users to investigate large gene sets and provides abilities to visualize semantic similarities, significant gene-disease associations, and gene set comparison. Funding was provided by the National Natural Science Foundation of China and Fundamental Research Funds for the Central Universities.DOSE is an R/Bioconductor package for semantic and enrichment analysis based on the Disease Ontology (DO). It enables biologists to explore disease similarities and gene function similarities in a disease context. DOSE supports semantic similarity computations between DO terms and genes, and enrichment analyses including hypergeometric model and gene set enrichment analysis (GSEA) to discover disease associations from high-throughput data. It also provides visualization functions for similarity and enrichment analyses, which are not available elsewhere. DOSE allows users to measure semantic similarity among DO terms and genes using several information-content and graph-structure based algorithms. It supports hypergeometric test and GSEA to evaluate disease relevance of biological experiments. DOSE provides several DO-specific visualization functions to produce highly customizable, publication-quality figures of similarity and enrichment analyses. The package includes functions for comparing disease associations among different gene sets. DOSE was developed using the R statistical computing language and is released within Bioconductor. It provides five algorithms for DO and gene semantic similarity measurements, hypergeometric test for identifying significant disease association of gene list, GSEA for interpreting genome-wide expression profiles in disease context, and comparison of significant disease associations among different gene sets. DOSE facilitates users to investigate large gene sets and provides abilities to visualize semantic similarities, significant gene-disease associations, and gene set comparison. Funding was provided by the National Natural Science Foundation of China and Fundamental Research Funds for the Central Universities.