[BUG] cdt.data.load_dataset('sachs') + one of the returned objects, 'target', is inconsistent with the paper(Sachs,etc 2005)

[xwbxxx]

The CDT version is 0.6.0 (as you can see in the first line of the output)

Problem: When I viewed the target, I found there exists a loop among PIP2, PIP3 and plcg（The top 3 nodes in my screenshot）.

As far as I know, Sachs Dataset should be a DAG, so I checked the paper where the graph came from, Causal-Protein-Signaling-Networks-Derived-from-Multiparameter-Single-Cell-Data(Sachs,etc 2005) to make sure. According to the paper, PIP3 should point to PIP2 . Moreover, there are a total of 20 edges in the diagram of the paper, but the function, cdt.data.load_dataset('sachs'), outputs a graph with only 18 edges. So I would like to know if the Sachs dataset is updated or if there are errors and missing edges in the graph from this function.

[diviyank]

Hello ! Thanks for the feedback ! I know that this dataset does have multiple versions (and revisions) ; I'll check and get back to you ! My version could be out-of-date

Best, Diviyan