The main obstacle to reasoning over the DOACC data is that there isn't an established body of theory to propose the kind of rules that enables useful inferences to be made. Merely collating the data is a necessary first step. However, once collated, the dataset can be explored for structural and emergent patterns.
A lot of preliminary spadework is required to get usable results. For example, in the below screenshot of Protege’s view of the DOACC dataset, the invitation to identify all the different individuals gives some indication of just how basic the knowledge modelling is. The performance of reasoners varies with the structure of the knowledge. Being excessively pedantic about the relationship between individuals to the excess of exhaustively listing every single individual as being different to every other individual, for every individual, yields a significant improvement in the reasoner’s performance.
Protege even provides a tool for that explicit purpose: