generate_ccans: Generate cis-co-accessibility networks (CCANs)

In cole-trapnell-lab/cicero-release: Predict cis-co-accessibility from single-cell chromatin accessibility data

Generate cis-co-accessibility networks (CCANs)

Description

Post process cicero co-accessibility scores to extract modules of sites that are co-accessible.

Usage

generate_ccans(
  connections_df,
  coaccess_cutoff_override = NULL,
  tolerance_digits = 2
)

Arguments

connections_df	Data frame of connections with columns: Peak1, Peak2, coaccess. Generally, the output of run_cicero or assemble_connections
coaccess_cutoff_override	Numeric, co-accessibility score threshold to impose. Overrides automatic calculation.
tolerance_digits	The number of digits to calculate cutoff to. Default is 2 (0.01 tolerance)

Details

CCANs are calculated by first specifying a minimum co-accessibility score and then using the Louvain community detection algorithm on the subgraph induced by excluding edges below this score. For this function, either the user can specify the minimum co-accessibility using coaccess_cutoff_override, or the cutoff can be calculated automatically by optimizing for CCAN number. The cutoff calculation can be slow, so users may wish to use the coaccess_cutoff_override after initially calculating the cutoff to speed future runs.

Value

Data frame with two columns - Peak and CCAN. CCAN column indicates CCAN assignment. Peaks not included in a CCAN are not returned.

Examples

## Not run: 
  data("cicero_data")
  set.seed(18)
  data("human.hg19.genome")
  sample_genome <- subset(human.hg19.genome, V1 == "chr18")
  sample_genome$V2[1] <- 100000
  input_cds <- make_atac_cds(cicero_data, binarize = TRUE)
  input_cds <- reduceDimension(input_cds, max_components = 2, num_dim=6,
                               reduction_method = 'tSNE',
                               norm_method = "none")
  tsne_coords <- t(reducedDimA(input_cds))
  row.names(tsne_coords) <- row.names(pData(input_cds))
  cicero_cds <- make_cicero_cds(input_cds, reduced_coordinates = tsne_coords)
  cicero_cons <- run_cicero(cicero_cds, sample_genome, sample_num = 2)
  ccan_assigns <- generate_ccans(cicero_cons)
 
## End(Not run)

cole-trapnell-lab/cicero-release documentation built on May 11, 2023, 11:12 p.m.