filter.dataset: Select (automatically) the most appropriate gene columns and...

In rlebron-bioinfo/gnlearn: Genetic Network Learning

Description

This function allows you to (automatically) select the most appropriate gene columns and cell rows of your dataset.

Usage

filter.dataset(
  df,
  filter.first = c("cells", "genes"),
  genes = NULL,
  selected.genes = NULL,
  features = NULL,
  max.genes = 100,
  max.cells = 2000,
  genes.criteria = c("non.zeros", "mean", "variance", "random"),
  cells.criteria = c("non.zeros", "mean", "variance", "random"),
  min.non.zeros = 10,
  cor = TRUE,
  cor.method = c("spearman", "kendall", "pearson"),
  cor.threshold = 0.25
)

Arguments

df	Dataset.
filter.first	Whether to filter the cells or the genes first. Default: cells
genes	Geneset with features (optional).
selected.genes	User-selected genes (optional).
features	Features you want to select (optional).
max.genes	Maximum number of selected genes (optional).
max.cells	Maximum number of selected cells (optional).
genes.criteria	Criteria to be applied when it is necessary to select a maximum number of genes from the previously filtered ones. Default: 'non.zeros'
cells.criteria	Criteria to be applied to select cells. Default: 'non.zeros'
min.non.zeros	Minimum number of non-zero values per gene (optional).
cor	Using correlation to exclude unconnected genes (optional).
cor.method	Correlation method: 'spearman', 'kendall', or 'pearson'. Default: 'spearman'
cor.threshold	Correlation threshold. Default: 0.1

Examples

df <- filter.dataset(df, max.genes=100, max.cells=2000)

rlebron-bioinfo/gnlearn documentation built on July 25, 2020, 12:38 p.m.