These dot files are generated from the files of RDF triples in ../ttl/ which are generated by the Monarch DIPper ingest scripts. The dot/image/ and dot/delta/image/ directories have bitmap images of the resulting graphs in case your browser does not know how to display the Graphviz dot format. The nodes in the graphs represent the name-spaces existing in the dataset. the edges in the graphs represent how and why names-paces are related. Hence (A)-b->(C) is stating that: Some identifier in 'A' is related to some identifier in 'C' by 'b' in this dataset. The number of relationships between name-spaces of a particular type is listed in parenthesis on the edge (Edge weight). Attempts have been made to provide human friendly labels on edges with opaque identifiers. The ./dot/delta/ directory has a combined dot file for every pair of dot files. (note before opening directory that is a lot of files, about 800 [summer 2016]) The dot files in the delta/ directory show what the graphs of the datasets have in common and how they differ. Red edges only exist in the graph of the first dataset. Green edges exist in the graphs of both datasets. Blue edges only exist in the graph of the second dataset. Counts (edge weights) are removed from the delta dot files as we only check for edge existence. Possible uses include: Checking test set coverage. (greener is better) for example zfin vs zfin_test. Assessing changes between different releases of the same dataset. Assessing the effect of changes to ingest process on the same dataset.