Using yaml to deal with gene sets in R

gene_list.r

install.packages("yaml")

# change path to your favorite local directory
gene_list <- function(filename, format = "yaml", path = "/Users/nacho/gene_info/gene_sets"){
    path <- file.path(path, filename)
    if (format == "yaml"){
        raw_genes <- yaml::yaml.load_file(path)
        # TODO: run through the nested names in genes and check that
        # there are no "." or ":", give warning or replace with "_"
        unlisted_genes <- unlist(raw_genes)
        groups <- sub("\\d+$", "", names(unlisted_genes))
        genes <- list(nested =  tapply(unlisted_genes, groups, unname))
        names(genes$nested) <- gsub('\\.', ':', names(genes$nested))
        genes$flat <- unique(unname(unlist(raw_genes)))
    } else {
        genes <- scan(path, "character")
    }
    genes
}

# example usage
my_isgs <- gene_list("isgs.yaml")$flat
[1] "CXCL10"    "CXCL11"    "DDX58" ...