R/multiple_sets_annot.R

In DanWiebe/fsgor: Package for Fold-specific GO Terms Recognition

#' Functional annotation for multiple gene sets
#'
#' Conducts functional annotation using Fisher exact test
#' for multiple gene sets, returns list with resulting annotations
#'
#' @param p_adjust_method method for multiple testing correction (to see all possible methods print: p.adjust.methods)
#'                        Benjamini-Hochberg by default
#' @param background_genes vector contains background set of genes
#' @param gene_amount_border minimal number of genes annotated to a term (1 by default)
#' @param gene_sets list contains quantile names (e.g. "1","2","1-6") as keys and corresponding gene sets as values
#' @param annot_data list object contains annotation data ready for functional annnotation procedure,
#'           ontology relation data, GO terms full names, GO terms namespaces and metadata for annotation
#'
#' @return list with filenames as keys and annotaton data frames as values
#' @export
#'
#' @examples
#'     data(up_genes, annot_data, package = "fsgor")
#'     gene_sets <- div_genes_to_quantiles(up_genes, 6)
#'     multiple_sets_annot(gene_sets, annot_data,  gene_amount_border = 10)
multiple_sets_annot <-
  function(gene_sets,
           annot_data,
           gene_amount_border = 1,
           background_genes = NULL,
           p_adjust_method = "BH") {

    annotlist <- annot_data$annot_data
    realnames <- annot_data$real_names
    namespaces <- annot_data$namespaces

    if (!is.null(background_genes)){
      annotlist <- fsgor::intersect_with_background(annotlist, background_genes)
    }

    outlist <- lapply(gene_sets, function(inputgenes) {

      # run functonal annotation
      return(
        func_annot_test_internal(
          annotlist,
          realnames,
          namespaces,
          inputgenes,
          gene_amount_border = gene_amount_border,
          p_adjust_method = p_adjust_method
        )
      )

    })

    names(outlist) <- names(gene_sets)
    return(outlist)
  }