TogoTable: cross-database annotation system using the Resource Description Framework (RDF) data model.

TogoTable (http://togotable.dbcls.jp/) is a web tool that adds user-specified annotations to a table that a user uploads. Annotations are drawn from several biological databases that use the Resource Description Framework (RDF) data model. TogoTable uses database identifiers (IDs) in the table as a query key for searching. RDF data, which form a network called Linked Open Data (LOD), can be searched from SPARQL endpoints using a SPARQL query language. Because TogoTable uses RDF, it can integrate annotations from not only the reference database to which the IDs originally belong, but also externally linked databases via the LOD network. For example, annotations in the Protein Data Bank can be retrieved using GeneID through links provided by the UniProt RDF. Because RDF has been standardized by the World Wide Web Consortium, any database with annotations based on the RDF data model can be easily incorporated into this tool. We believe that TogoTable is a valuable Web tool, particularly for experimental biologists who need to process huge amounts of data such as high-throughput experimental output.

Detection of p53 mutations in proliferating vascular cells in glioblastoma multiforme.

OBJECT: Glioblastoma multiforme (GBM), one of the most aggressive tumors in humans, is highly angiogenic. However, treatment with the angiogenesis inhibitor bevacizumab has not significantly prolonged overall patient survival times. GBM resistance to angiogenesis inhibitors is attributed to multiple interacting mechanisms. Although mesenchymal transition via glioma stem-like cells has attracted attention, it is considered a poor biomarker. There is no simple method for differentiating tumor-derived and reactive vascular cells from normal cells. The authors attempted to detect the mesenchymal transition of tumor cells by means of p53 and isocitrate dehydrogenase 1 (IDH1) immunohistochemistry. METHODS: Using antibody against p53 and IDH1 R132H, the authors immunohistochemically analyzed GBM tissue from patients who had undergone surgery at the University of Miyazaki Hospital during August 2005-December 2011. They focused on microvascular proliferation with a p53-positive ratio exceeding 50%. They compared TP53 mutations in original tumor tissues and in p53-positive and p53-negative microvascular proliferation cells collected by laser microdissection. RESULTS: Among 61 enrolled GBM patients, the first screening step (immunostaining) identified 46 GBMs as p53 positive, 3 of which manifested areas of prominent p53-positive microvascular proliferation (>50%). Histologically, areas of p53-positive microvascular proliferation tended to be clustered, and they coexisted with areas of p53-negative microvascular proliferation. Both types of microvascular proliferation cells were clearly separated from original tumor cells by glial fibrillary acidic protein, epidermal growth factor receptor, and low-/high-molecular-weight cytokeratin. DNA sequencing analysis disclosed that p53-positive microvascular proliferation cells exhibited TP53 mutations identical to those observed in the original tumor; p53-negative microvascular proliferation cells contained a normal allele. Although immunostaining indicated that 3 (2 primary and 1 secondary) of the 61 GBMs were positive for IDH1, no tumors contained microvascular proliferation cells positive for IDH1 R132H. CONCLUSIONS: Some microvascular proliferation clusters in GBM result from mesenchymal transition. The identification of useful markers might reveal this phenomenon as an infrequent event in GBMs.