Multiple edges for Post Analysis

[mikekucera]

But could we have multiple edges for different runs of PA? currently we can have multiple edges but that is run for a different dataset (and the type is sig os sig_set2 on the edge). In this scenario it would be the exact same edge just with a different p-value for the mann-whitney.

Not sure about edge color. Might be too busy. Here is an example using color. Remember that if there are two datasets the lines would be either solid or dashed. Veronique, do you think it is too messy?

[mikekucera]

Can we maybe have options:
lesser - compute just lesser than edges (make all created edge color for lesser)
greater - compute just greater than edges (make all created edge color for greater)
two.sided - - compute just two.sided than edges (make all created edge color for two.sided)
All - being where you create multiple edges, one for each of the above three tests (only if they pass the threshold)
less/greater - only compute lesser and greater

Currently edges are mapped to Enrichment_set. could we change that Test (or something like that).
for EM edges test = jaccard/overlap/combined
for PA edges test = hypergeometrix/two.sided/lesser/greater mann.whitney