Build a weighted feature correlation network
Usage
BuildFeatureWGCNA(
data,
variables,
sample_id = NULL,
network_type = c("signed", "unsigned", "signed_hybrid"),
soft_power = NULL,
sft_rsq = 0.85,
min_module_size = 30,
deep_split = 2,
merge_cut_height = 0.25,
keep_tom = FALSE,
seed = NULL
)Arguments
- data
A data frame with samples in rows and features in columns.
- variables
Explicit feature columns to include.
- sample_id
Optional unique sample-ID column retained for trait joins.
- network_type
One of
"signed","unsigned", or"signed_hybrid".- soft_power
Optional soft-thresholding power.
- sft_rsq
Target scale-free topology fit for automatic power selection.
- min_module_size
Minimum dynamic-tree-cut module size.
- deep_split
Dynamic tree-cut sensitivity from 0 through 4.
- merge_cut_height
Eigengene dissimilarity for module merging.
- keep_tom
Store the topological-overlap matrix.
- seed
Optional random seed.