sccore: smooth_count_matrix – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

smooth_count_matrix

Smooth gene expression, used primarily within conos::correctGenes. Used to smooth gene expression values in order to better represent the graph structure. Use diffusion of expression on graph with the equation dv = exp(-a * (v + b))

Description

Usage

smooth_count_matrix(
  edge_verts,
  edge_weights,
  count_matrix,
  is_label_fixed,
  max_n_iters = 10L,
  diffusion_fading = 1,
  diffusion_fading_const = 0.1,
  tol = 0.001,
  verbose = TRUE,
  normalize = FALSE
)

Arguments

`edge_verts`	edge vertices of igraph graph object
`edge_weights`	edge weights of igraph graph object
`count_matrix`	gene count matrix
`is_label_fixed`	boolean Whether label is fixed
`max_n_iters`	integer Maximal number of iterations (default=10)
`diffusion_fading`	numeric Constant used for diffusion on the graph, exp(-diffusion.fading * (edge_length + diffusion.fading.const)) (default=1.0)
`diffusion_fading_const`	numeric Another constant used for diffusion on the graph, exp(-diffusion.fading * (edge_length + diffusion.fading.const)) (default=0.1)
`tol`	numeric Absolute tolerance as a stopping criteria (default=1e-3)
`verbose`	boolean Verbose mode (default=TRUE)
`normalize`	boolean Whether to normalize values (default=FALSE)

Value

matrix from input graph, with labels propagated

sccore

Core Utilities for Single-Cell RNA-Seq

v0.1.3

GPL-3

Authors

Viktor Petukhov [aut], Ramus Rydbirk [aut], Peter Kharchenko [aut], Evan Biederstedt [aut, cre]

Initial release