Include count of number of genes detected

[dosumis]

Not sure whether this can be done easily here or downstream using cypher. If latter, this query works:

MATCH (gene:Gene)<-[:expresses]-(clu:Cluster)-[:has_source]->(ds:DataSet)
WITH ds, apoc.coll.toSet(COLLECT(gene)) AS genes 
RETURN ds.label as dataset, size(genes) AS depth order by depth DESC

Think high priority to get this in and to expose through API. Users need a way to choose the highest seq depth dataset available for their cell type as it makes a big difference to the results.

[Clare72]

Yes, can be done here. Do we want number of genes for any extent (pre-filtering) or just extent >0.2 (post-filtering). If post-filtering, it might be easier to do in cypher (but impossible to do in cypher for pre-filtering numbers).

[dosumis]

Do we want number of genes for any extent (pre-filtering) or just extent >0.2 (post-filtering).

Good point. I think post filtering is what's needed to inform queries, but might be useful to have both.

[Clare72]

gene counts added to clusters and datasets - now in files in e809bf3
need to update vfb-connect to use these - see VirtualFlyBrain/VFB_connect#224