also been tested. Other characteristics observed involve: representation of the annotation (intrusive - in the text or non-intrusive - as attached file), types of documents compatible (which file extensions are supported as an input in the tool), documentation availability and platform of development (web or desktop). Despite the fact that tools such as KIM, Ontea and RDFace generate automatic annotations, they have their own ontologies which are not on the biomedical domain. Knowtator is a plugin for Protégé Server and only a few tasks of the annotation process are automated. MnM, GoNTogle and RDFa Editor tools perform automatic annotation and have flexibility in loading arbitrary ontologies, but could not be used due to support problems. The NCBO Annotator is a web service that annotates full texts using ontologies from biomedical domain available at NCBO BioPortal. However, it is not available for immediate usage, and demands the development of a client to that web service. AutôMeta and GATE can perform automatic annotation of documents and also have flexibility in loading arbitrary ontologies. These are selected as tools to be tested and used for the purpose of our experiment on semantic annotation. AutôMeta uses RDFa (an annotation language recommended by W3C), has a reasoner to infer new annotations and it supports the load of large ontologies such as Gene Ontology and NCI Thesaurus, among others. The texts for annotation must be in 'txt' format and annotation is made using an intrusive method. GATE is a tool for natural language processing. It is very solid and mature in the task of semantic annotation using the resources of language and processing. Its differential is on being able to load different extensions of documents (txt, pdf, doc, etc.). Additionally, it performs non-intrusive annotation, and archives them in 'xml' files. Ontologies are loaded as processing resources, which can happen very slowly in the case of large ontologies. Both tools have good documentations and are free. Moreover, it is noteworthy that these tools works with a set of input texts, but only uses one ontology at a time. Therefore, it is possible to have texts annotated with multiple ontologies, but in separate files, generating a new volume of texts and many output files. The simultaneous annotation with multiple ontologies is still an unsolved problem.