Evaluates expressions against the edge table and stores the results as edge columns.
Arguments
- x
Network input.
- ...
Named expressions evaluated against the edge table, with the same metrics and predicates
select_edges()offers, for examplestrong = abs_weight > 0.5orscaled = weight / max(weight).- community
Community detection method used when an expression refers to
same_community,from_communityorto_community. Default"louvain".- keep_format
Logical. Return the input format when TRUE.
- directed
Logical or NULL. If NULL (default), auto-detect.
Value
A cograph_network whose edge table has the new columns, or
the input format when keep_format = TRUE.
Examples
as.data.frame(mutate_edges(regulation_net, strong = weight > 0.2))
#> from to weight strong
#> 1 Adapt Explore 0.28 TRUE
#> 2 Reflect Explore 0.05 FALSE
#> 3 Discuss Explore 0.30 TRUE
#> 4 Create Explore 0.14 FALSE
#> 5 Synthesize Plan 0.11 FALSE
#> 6 Share Plan 0.21 TRUE
#> 7 Plan Monitor 0.13 FALSE
#> 8 Reflect Monitor 0.15 FALSE
#> 9 Synthesize Monitor 0.07 FALSE
#> 10 Evaluate Monitor 0.33 TRUE
#> 11 Create Monitor 0.17 FALSE
#> 12 Share Monitor 0.49 TRUE
#> 13 Monitor Adapt 0.16 FALSE
#> 14 Evaluate Adapt 0.43 TRUE
#> 15 Share Adapt 0.39 TRUE
#> 16 Explore Reflect 0.35 TRUE
#> 17 Discuss Reflect 0.35 TRUE
#> 18 Synthesize Reflect 0.42 TRUE
#> 19 Evaluate Reflect 0.07 FALSE
#> 20 Plan Discuss 0.40 TRUE
#> 21 Adapt Discuss 0.34 TRUE
#> 22 Adapt Synthesize 0.17 FALSE
#> 23 Plan Evaluate 0.49 TRUE
#> 24 Create Evaluate 0.39 TRUE
#> 25 Plan Create 0.20 FALSE
#> 26 Monitor Create 0.37 TRUE
#> 27 Discuss Create 0.14 FALSE
#> 28 Explore Share 0.27 TRUE
#> 29 Plan Share 0.36 TRUE
#> 30 Create Share 0.23 TRUE
