select.genes: Select (automatically) the most appropriate gene columns of...
In rlebron-bioinfo/gnlearn: Genetic Network Learning

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

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

`df`	Dataset.
`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).
`selection.criteria`	Criteria to be applied when it is necessary to select a maximum number of genes from the previously filtered ones. 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

1 2	df <- select.genes(df, max.genes=100, min.non.zeros=2) df <- select.genes(df, genes=genes, features=c('tumor.suppressor', 'breast.cancer'), cor=TRUE, cor.threshold=0.7)