cleanData: Filter sparse matrix

In plantformatics/Socrates: Analysis of scATAC-seq data

Filter sparse matrix

Description

This function filters the cell x peak matrix based on pre-clustering results, a minimum peak usage frequency, and number of accessible peaks per cell

Usage

cleanData(
  obj,
  preclusterID = NULL,
  min.p = 0.01,
  min.t = 0.01,
  min.c = 100,
  max.t = 0.005,
  min.cells.precluster = 10,
  verbose = F
)

Arguments

obj	list (counts and meta) output from loadSparseData
preclusterID	character column name in meta data for LSI or other clustering results to condition ACR peak frequency. Sets the minimum frequency of accessibility across cells within a single cluster, where sites that pass this threshold in at least one cluster are retained. Defaults to NULL.
min.p	numeric, minimum frequency of accessible cells per cluster (see preclusterID). Sites above this threshold in at least one cluster are retained. Acceptable values range from 0 to 1. Defaults to 0.01.
min.t	filters ACRs below the specified frequency of accessible cells across all cells. Acceptable values range from 0 to 1.
min.c	minimum number of accessible ACRs per cell. Defaults to 100.
max.t	remove top X% of ACRs by frequency across cells (remove constitutive accessible regions). Defaults to 0.005
min.cells.precluster	minimum number cells from precluster (preclusterID) to include cluster for thresholds set by min.p. Defaults to 10
verbose	Defaults to FALSE

plantformatics/Socrates documentation built on April 3, 2025, 1:02 p.m.