Abstract. Skeletal dysplasias comprise a group of genetic diseases characterized by highly complex, heterogeneous and sparse data. Performing efficient and automated knowledge discovery in this domain poses serious challenges, one of the main issues being the lack of a proper formalization. Semantic Web technologies can, however, provide the appropriate means for encoding the knowledge and hence enabling complex forms of reasoning. We aim to develop decision support methods in the skeletal dysplasia domain by applying uncertainty reasoning over Semantic Web data. More specifically, we devise techniques for semi-automated diagnosis and key disease feature inferencing from an existing pool of patient cases -that are shared and discussed in the SKELETOME community-driven knowledge curation platform. The outcome of our research will enable clinicians and researchers to acquire a critical mass of structured knowledge that will sustain a better understanding of these genetic diseases and foster advances in the field.