Biological ontologies are widely used for genome annotation. Identifying correspondences between concepts of two ontologies (mapping) allows the reuse and sharing of annotations. Accordingly, biological ontology mapping has attracted a lot of interest. In this paper, we introduce O’Browser, a semi-automatic method for mapping two functional hierarchies using two sets of carefully annotated proteins. While being based on a classical ontology mapping architecture, O’Browser computes correspondences using a combination of different kinds of matchers. A key feature of O’Browser is that it places the expert at the center of the mapping process at two stages: (i) both to validate the very strong correspondences discovered by the system and to identify functional groups of concepts and (ii) to validate the correspondences given by the combination of results found by the matchers. These matchers have been designed in O’Browser to fit best with functional hierarchy features. For instance, we have introduced a new instance-based matcher which uses homology relationships between proteins. The combination of the different matchers is based on an original notion of adaptive weighting. Here, we show the ability of O’Browser to map concepts of Subtilist to concepts of FunCat, two functional hierarchies. First results appear to be very promising.