In the available data download files, you have provided data as Associations direct and Indirect respectively, but they also contain data that is divided into direct and indirect.
For example, I would like to know what you mean by “indirect data” included in “associations direct”.
direct associations refer to the aggregation of all evidence for a fixed disease and target identifiers. The indirect associations dataset is the result of expanding the disease ID that was originally reported to its parents terms in the EFO ontology. You can read more about it in our docs.
I don’t understand what kind of reference to indirect data you are finding in the direct dataset. These are for example the available fields in the overall direct associations dataset.
|-- diseaseId: string (nullable = true)
|-- targetId: string (nullable = true)
|-- score: double (nullable = true)
|-- evidenceCount: long (nullable = true)
Could you please illustrate your question with an example?