Database Persistence

[chrtannus]

For each successful query, save into MongoDB:

• The network, including:
  • network ID
  • network name
  • network topology¹
  • node and edge attributes¹
  • node positions
• The iRegulon results (motifs and tracks with their target genes and TFs, ranks, NES values, etc.) -- I don't think we need to save the TF view, since it's derived from the motifs and tracks.
• Motifs/Tracks/Clusters selection, including the currently selected TFs (from the data tables).
• The query parameters -- see Query iRegulon - predict regulators and targets #2 and Query the Metatargetome #3
• The creation time and last access time

NOTE: I think the demo network doesn't need to be stored in Mongo. Just create it again every time from static or even in-memory data. Or if it's more convenient to load the demo data from the same DB, the network data does not need to be updated (e.g. with node positions) -- see #8.

Footnotes

1. Maybe we don't need to save the network elements, because the network can be recreated from the results again when reloading the data. Either way, the persisted data must contain information about which Motifs/Tracks/TFs were used to create the network--we also need this information to restore the checked rows when loading the data tables. ↩ ↩²