scran_graph_cluster
Graph-based clustering of cells
|
Options for cluster_walktrap()
.
More...
#include <cluster_walktrap.hpp>
Public Attributes | |
int | steps = 4 |
bool | report_merges = true |
bool | report_modularity = true |
Options for cluster_walktrap()
.
int scran_graph_cluster::ClusterWalktrapOptions::steps = 4 |
Number of steps of the random walk. This determines the ability of the Walktrap algorithm to distinguish highly interconnected communities from the rest of the graph. Ideally, a random walk of the specified length from a node should easily reach any other node in the same community but not a node of a different community (i.e., it is "trapped" in the same community, hence the name of the method). The default is based on the example in the igraph documentation.
bool scran_graph_cluster::ClusterWalktrapOptions::report_merges = true |
Whether to report the merge steps in Results::merges
.
bool scran_graph_cluster::ClusterWalktrapOptions::report_modularity = true |
Whether to report the modularity after each merge step in Results::modularity
.