R/get_rank_tables.R

In scMAGeCK: Identify genes associated with multiple expression phenotypes in single-cell CRISPR screening data

# virtura_facs_functions
get_rank_tables <- function(genes_to_rank, negctrlgenelist = "NonTargetingControlGuideForHuman") {
    score_test = seq(-1, 1, length.out = length(genes_to_rank))
    score_tggene = paste(genes_to_rank, 1:length(genes_to_rank), sep = "_")
    names(score_test) = score_tggene
    
    candidate_gene = unique(genes_to_rank)
    candidate_gene = candidate_gene[!is.na(candidate_gene)]
    gsea_score = rep(1, length(candidate_gene))
    p_vsall = gsea_score
    p_vsnegctrl = gsea_score
    for (tg_i in 1:length(candidate_gene)) {
        
        tggene = candidate_gene[tg_i]
        # message(paste(tggene,'...'))
        ghit = score_tggene[genes_to_rank == tggene]
        
        gs_score = getoriginalgseascore(score_test, ghit)
        z = max(gs_score)
        gsea_score[tg_i] = z
        
        z = ks.test(score_test[!score_tggene %in% ghit], score_test[ghit])
        p_vsall[tg_i] = z$p.value
        
    
        
        z = ks.test(score_test[genes_to_rank %in% negctrlgenelist], score_test[ghit])
        p_vsnegctrl[tg_i] = z$p.value
        
    }
    
    report_f = data.frame(gene = candidate_gene, gsea = gsea_score, ks_p = p_vsall, ks_negctrl = p_vsnegctrl)
    
    report_f[, "ks_p_adj"] = p.adjust(report_f$ks_p, method = "fdr")
    
    report_f[, "ks_negctrl_adj"] = p.adjust(report_f$ks_negctrl, method = "fdr")
    return(report_f)
}
TRUE