Use Leiden algorithm as implemented in igraph>=1.2.7 #378
Labels
Comments
GeorgescuC
added a commit
that referenced
this issue
Jul 6, 2022
…mentation from igraph rather than that of the Leiden package that requires reticulate and a python install. This reduces dependencies and avoid the recent errors and warnings elevated to errors from the Leiden package. Should also be faster as originally suggested in #378 by @vtraag . Replace reshape with reshape2 for melt() import as the first raises warnings in newer R versions.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
In igraph/python-igraph#458 we realised that
inferCNVis using theleidenR package that usesreticulateto use myleidenalgPython package.As of version 1.2.7 the Leiden algorithm is also implemented directly in
igraph, see igraph/rigraph#346. That implementation should be faster, and more straightforward to use and doesn't requirereticulateor Python. The downside is that it doesn't offer the same degree of flexibility (e.g. no multilayer support, only supports CPM/Modularity), but if you don't depend on these additional options, it may be easier to switch to usingcluster_leidenas implemented inigraph.The text was updated successfully, but these errors were encountered: