Spatial modeling of interactions between APCs, effector and regulatory T cells.

Not Supplied.

ZF-HDD: a database for zebrafish models of Human disease.

Not Supplied.

Relating Biological Information through Literature.

The mining of biological literature is an important recent research topic, motivated by the large number of biological articles that curators have to read in order to update biological databases, or simply to be aware of progress in a specific area. ReBIL (Relating Biological Information through Literature) aims at providing text-mining tools for biological literature that avoid the complex issues of creating rules and patterns encompassing all possible cases and training sets that are too specific to be extended to new domains. ReBIL has been developing the following Web tools:

- GOAnnotator (http://xldb.fc.ul.pt/rebil/tools/goa/) for verification of electronic protein annotations using GO terms automatically extracted from literature;

- FuSSiMeG (http://xldb.fc.ul.pt/rebil/tools/ssm/) provides a functional similarity measure between two proteins using the semantic similaritybetween the GO terms annotated with the proteins;

- WebAPEG (http://xldb.fc.ul.pt/rebil/tools/apeg/) provides functional annotations automatically extracted from literature of genes from the Arabidopsis Pollen Expressed Gene database;

- ProFAL (bioProducts Functional Annotation through Literature) is a system for automatic annotation of biological databases using the following methods:

- WeBTC (Web Biological Text Classification) is a novel method for text classification on biomedical literature, involving the use of information extracted from related web resources.

- FiGO (Finding Genomic Ontology) is a novel unsupervised method for identifying biological properties organized in a genomic ontology in unstructured text using the information content of each word present in the nomenclature of the ontology.

- CAC (Correlate the Annotations' Components) is a novel method for discarding misannotations identified by automated systems using curated annotations with similar structure and function.

- GraSM (Graph-based Similarity Measure) is a novel similarity measure that incorporates the semantic richness of a graph into a semantic similarity measure, instead of just using an ontology as a tree-like hierarchy.