search

meringlab / FlashWeave.jl

Inference of microbial interaction networks from large-scale heterogeneous abundance data
Other
70 stars 8 forks source link

Interpreting track_rejections output #35

Closed michmmcknight closed 8 months ago

michmmcknight commented 9 months ago

I am wanting to explore the discarded edges from my network to determine which ones are the result of environmental variables and included the generations of a track_rejections file when running flashweave (see head of file below). I've gathered that the Edge column indicates the ID of the two nodes connected by the rejected edge, but am looking for clarification on the values in the Rejecting_set column. Is it correct to assume the numbers there also representative of node IDs? Any clarifications appreciated!

Edge Rejecting_set Stat P_value Num_tests Perc_tested Df SuffPower 4986 <-> 4657 1400,3067,1106 0.0 1.0 1 0.14286 0 true 4986 <-> 4805 1400,3067,1106 0.0 1.0 1 0.14286 0 true 4986 <-> 3626 1400,3067,1106 0.0 1.0 1 0.14286 0 true 4986 <-> 4636 1400,3067,1106 0.0 1.0 1 0.14286 0 true 4986 <-> 2802 1400,3067,1106 0.0 1.0 1 0.14286 0 true

jtackm commented 9 months ago

Yes, you are correct. Did you provide meaningful OTU headers in your data files or to the 'learn_network' function? If yes, and the track_rejections file still only shows numerical node IDs, that would be a bug to fix.