R/plotMutProfile.R

In MesKit: A tool kit for dissecting cancer evolution from multi-region derived tumor biopsies via somatic alterations

#' @title plotMutProfile
#'
#' @param maf Maf or MafList object generated by \code{\link{readMaf}} function.
#' @param patient.id  Select or reorder the patients. Default NULL, all patients are included.
#' Classify SSNVs/Indels into Shared/P-shared/Private, Clonal/Subclonl
#' or Shared-Clonal/P-shared-Clonal/Private-Clonal/Shared-Subclonal/P-shared-SubClonal/Private-SubClonal 
#' @param class  The class which would be represented. Default "SP" (Shared pattern: Public/Shared/Private),
#' other options: "CS" (Clonal status: Clonal/Subclonl) and "SPCS".
#' @param classByTumor  Logical (Default: FALSE). Define shared pattern of mutations based on tumor types (TRUE) or samples (FALSE)
#' @param topGenesCount  The number of genes print, Default 10.
#' @param geneList  A list of genes to restrict the analysis. Default NULL.
#' @param sample.text.size Fontsize of sample name. Default 11.
#' @param gene.text.size Fontsize of gene text. Default 11.
#' @param legend.text.size Fontsize of legend text. Default 11.
#' @param legend.title.size Fontsize of legend title. Default 11.
#' @param patientsCol  A list containing customized colors for distinct patients. Default NULL.
#' @param bgCol  Background grid color. Default "#f0f0f0".
#' @param removeEmptyCols  Logical (Default: TRUE). Whether remove the samples without alterations.
#' @param removeEmptyRows  Logical (Default: TRUE). Whether remove the genes without alterations.
#' @param showColnames  Logical (Default: TRUE). Show sample names of columns.
#' @param sampleOrder A named list which contains the sample order used in plotting the final profile. Default NULL.
#' @param use.tumorSampleLabel Logical (Default: FALSE). Rename the 'Tumor_Sample_Barcode' with 'Tumor_Sample_Label'.
#' @param ... Other options passed to \code{\link{subMaf}}
#' @return Mutational profile
#' 
#' @examples
#' maf.File <- system.file("extdata/", "CRC_HZ.maf", package = "MesKit")
#' clin.File <- system.file("extdata/", "CRC_HZ.clin.txt", package = "MesKit")
#' ccf.File <- system.file("extdata/", "CRC_HZ.ccf.tsv", package = "MesKit")
#' maf <- readMaf(mafFile=maf.File, clinicalFile = clin.File, ccfFile=ccf.File, refBuild="hg19")
#' plotMutProfile(maf, class = "SP")
#' @import ComplexHeatmap
#' @importFrom stats na.omit
#' @export plotMutProfile


plotMutProfile <- function(maf,
                           patient.id = NULL,
                           class = "SP",
                           classByTumor = FALSE,
                           topGenesCount = 10,
                           geneList = NULL,
                           sample.text.size = 11,
                           gene.text.size = 11,
                           legend.text.size = 11,
                           legend.title.size = 11,
                           bgCol = "#f0f0f0",
                           patientsCol = NULL,
                           removeEmptyCols = TRUE,
                           removeEmptyRows = TRUE, 
                           showColnames = TRUE,
                           sampleOrder = NULL,
                           use.tumorSampleLabel = FALSE,
                           ...) {
  
  ## filter maf and order patient
  maf_input <- subMaf(maf,
                      patient.id = patient.id,
                      mafObj = TRUE,
                      use.tumorSampleLabel = use.tumorSampleLabel,
                      ...)
  if (any(names(maf_input) != patient.id)) {
    mafTemp <- list()
    for (i in seq_len(length(maf_input))) {
      mafTemp[i] <- maf_input[which(names(maf_input) == patient.id[i])]
    }
    names(mafTemp) <- patient.id
    maf_input <- mafTemp
  }
  
  
  ## merge maf data
  maf_data_list <- lapply(maf_input, getMafData)
  
  # order samples
  if (!(is.null(sampleOrder))) {
    for (i in seq_len(length(sampleOrder))) {
      if (names(sampleOrder)[i] %in% patient.id) {
        
        # select related maf data 
        mafRelated <- maf_data_list[names(sampleOrder)[i]]
        if (all(sampleOrder[[i]] %in% unique(mafRelated[[1]]$Tumor_Sample_Barcode))) {
          
          # filter samples based on sampleOrder
          mafFiltered <- mafRelated[[1]][which(mafRelated[[1]]$Tumor_Sample_Barcode %in% sampleOrder[[i]]), ]
          
          # order samples 
          mafFiltered$Tumor_Sample_Barcode <- factor(mafFiltered$Tumor_Sample_Barcode, levels = sampleOrder[[i]])
          mafOrdered <- mafFiltered[with(mafFiltered, order(mafFiltered$Tumor_Sample_Barcode)), ]
          mafOrdered$Tumor_Sample_Barcode <- as.character(mafOrdered$Tumor_Sample_Barcode)
          
        } else {
          stop(paste0("sampleOrder should be consistent with Tumor_Sample_Barcode in Maf obejct."))
        }
      }
      maf_data_list[names(sampleOrder)[i]][[1]] <- as.data.table(mafOrdered)
    }
  }
  
  maf_data <- maf_data_list[[1]]
  
  for (d in maf_data_list[-1]){
    maf_data <- rbind(maf_data, d)
  }
  maf_data<- as.data.frame(maf_data)
  
  maf_data <- do.classify(maf_data, classByTumor = classByTumor, class = class)
  
  patient.split <- maf_data %>%
    dplyr::select("Patient_ID", "Tumor_Sample_Barcode") %>%
    dplyr::distinct() %>%
    dplyr::select("Patient_ID") %>%
    as.matrix() %>%
    as.vector() %>%
    as.character()
  
  maf_data_before <- maf_data
  
  if (!is.null(geneList)) {  
    
    maf_data <- maf_data %>%
      dplyr::rowwise() %>%
      dplyr::mutate(Selected_Mut = dplyr::if_else(
        any(.data$Hugo_Symbol %in% geneList),
        TRUE,
        FALSE)) %>%
      dplyr::filter(.data$Selected_Mut)
  }
  
  
  
  if(length(unique(patient.split)) == 1){
    patient.split = NULL
  }
  # long -> wider
  mat <- maf_data %>%
    dplyr::ungroup() %>%
    dplyr::group_by(.data$Hugo_Symbol) %>%
    dplyr::mutate(
      total_barcode_count = sum(.data$unique_barcode_count)
    ) %>%
    dplyr::select("Hugo_Symbol",
                  "Patient_ID",
                  "Tumor_Sample_Barcode",
                  "Mutation_Type",
                  "total_barcode_count"
    ) %>%
    tidyr::pivot_wider(
      #names_from = Tumor_Sample_Barcode,
      names_from = c("Patient_ID", "Tumor_Sample_Barcode"),
      values_from = "Mutation_Type",
      names_sep = ":",
      values_fn = list("Mutation_Type" = multiHits)
    ) %>%
    dplyr::ungroup() %>%
    dplyr::arrange(dplyr::desc(.data$total_barcode_count))      
  #dplyr::select_if(function(x) {!all(is.na(x))}) %>%
  
  ## add NA columns for filtered samples, if exist
  ori_sample <- unique(paste(maf_data_before[, 'Patient_ID'], 
                             maf_data_before[, 'Tumor_Sample_Barcode'], 
                             sep = ":"))
  if(ncol(mat) - 2 != length(ori_sample)){
    helper_dat <- as.data.frame(matrix(ncol = length(ori_sample),
                                       nrow = nrow(mat)))
    colnames(helper_dat) <- ori_sample
    for (i in colnames(helper_dat)) {
      if (i %in% colnames(mat)){
        helper_dat[, i] <- mat[, i]
      }
    }
    
    mat <- cbind(mat[, 1:2], helper_dat)
  }
  
  
  if (nrow(mat) < topGenesCount) {
    message(paste0("Warning: only ", nrow(mat), ' genes was/were found in this analysis.'))
  } else{
    
    mat <- as.data.frame(mat) %>% dplyr::slice(seq_len(topGenesCount))
    
    #tibble::column_to_rownames(., "Hugo_Symbol") %>% 
    
    matTemp <- mat[, (seq_len(ncol(mat) - 1) + 1)] 
    rownames(matTemp) <- mat$Hugo_Symbol
    
    mat <- matTemp %>% 
      dplyr::select(-"total_barcode_count") %>% as.matrix()
    
  }
  
  col_labels <- dplyr::select(maf_data_before, "Patient_ID", "Tumor_Sample_Barcode")%>%
    dplyr::distinct()
  col_labels <- as.vector(col_labels$Tumor_Sample_Barcode)
  
  # get the order of rows
  stat <- rep(0, topGenesCount)
  for(i in seq_len(nrow(mat))){
    stat[i] <- sum(!is.na(mat[i, ])) / ncol(mat)
  }
  
  rowOrderFrame <- data.frame(Genes = rownames(mat), freq = stat)
  rowOrder <- as.numeric(rownames(rowOrderFrame[order(rowOrderFrame$freq, decreasing = TRUE), ]))
  
  # View(mat)
  
  #patient_id_cols <-
  #RColorBrewer::brewer.pal(length(unique(patient.split)), "Set")
  #names(patient_id_cols) <- unique(patient.split)
  #
  #patient_barcode <- maf_data %>%
  #dplyr::select(Patient_ID, Tumor_Sample_Barcode) %>%
  #dplyr::distinct() %>%
  #dplyr::mutate(color = patient_id_cols[patient.split]) 
  #
  #sample_barcode <- patient_barcode$color
  #names(sample_barcode) <- patient_barcode$Tumor_Sample_Barcode
  
  
  multi_hit_exist = FALSE
  for (i in seq_len(nrow(mat))) {
    for (j in seq_len(ncol(mat))) {
      if (length(temp <- grep("Multi_hits", mat[i, j]))) {
        multi_hit_exist = TRUE
        break
      }
    }
  }
  
  if (!(classByTumor)) {
    col_type <- function(class) {
      if (class == "SP") {
        cols <- c("#3C5488FF", "#00A087FF", "#F39B7fFF")
        names(cols) <- c("Public","Shared", "Private")
      } else if (class == "CS") {
        cols <- c("#00A087FF", "#3C5488FF")
        names(cols) <- c("Clonal", "Subclonal")
      } else if (class == "SPCS") {
        cols <-
          c(
            "#00A087FF",
            "#3C5488FF",
            "#8491B4FF",
            "#F39B7FFF", 
            "#E64B35FF",                    
            "#4DBBD5FF"                    
          )
        names(cols) <-
          c(
            "Public_Clonal",
            "Public_Subclonal",
            "Shared_Clonal",
            "Shared_Subclonal",
            "Private_Clonal",
            "Private_Subclonal"                    
          )
        
      }
      mutTypes <- as.character(stats::na.omit(unique(unlist(strsplit(mat, ";")))))
      cols <- cols[which(names(cols) %in% mutTypes)]
      
      return(cols)
    }
    
    colorSelect <- col_type(class)
    
    alter_fun <- function(class){
      
      if(class == "SP"){
        l <- list(
          #background = function(x, y, w, h)
          #grid::grid.rect(x, y, w * 0.9, h * 0.9,
          # gp = grid::gpar(fill = bgCol, col = NA)),
          Private = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Private"], col = NA)),
          Public = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Public"], col = NA)),
          Shared = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Shared"], col = NA)),
          Multi_hits = function(x, y, w, h)
            grid::grid.points(x, y, pch = 16, size = grid::unit(0.5, "char")
            ))
      }else if(class == "CS"){
        l <- list(
          #background = function(x, y, w, h)
          #grid::grid.rect(x, y, w * 0.9, h * 0.9,
          #gp = grid::gpar(fill = bgCol, col = NA)),
          Clonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Clonal"], col = NA)),
          Subclonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Subclonal"], col = NA)),
          Multi_hits = function(x, y, w, h)
            grid::grid.points(x, y, pch = 16, size = grid::unit(0.5, "char") 
            ))
      }else if(class == "SPCS" ){
        l <- list(
          #background = function(x, y, w, h)
          #grid::grid.rect(x, y, w * 0.9, h * 0.9,
          #gp = grid::gpar(fill = bgCol, col = NA)),
          Private_Clonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Private_Clonal"], col = NA)),
          Private_Subclonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Private_Subclonal"], col = NA)),
          Public_Clonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Public_Clonal"], col = NA)),
          Public_Subclonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Public_Subclonal"], col = NA)),
          Shared_Clonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Shared_Clonal"], col = NA)),
          Shared_Subclonal = function(x, y, w, h)
            grid::grid.rect(x, y, w * 0.9, h * 0.9,
                            gp = grid::gpar(fill = colorSelect["Shared_Subclonal"], col = NA)),
          Multi_hits = function(x, y, w, h)
            grid::grid.points(x, y, pch = 16, size = grid::unit(0.5, "char") 
            ))
      }
      
      mutTypes <- as.character(stats::na.omit(unique(unlist(strsplit(mat, ";")))))
      l_filter <- l[which(names(l) %in% mutTypes)]
      l_final <- c(background = function(x, y, w, h)
        grid::grid.rect(x, y, w * 0.9, h * 0.9,
                        gp = grid::gpar(fill = bgCol, col = NA)), l_filter)
      
      
      return(l_final)            
    }
    
    
    
  } else{
    
    # set certain colors
    colorScale <- c("#3C5488FF", "#00A087FF", "#F39B7fFF",
                    "#8491B4FF","#E64B35FF","#4DBBD5FF",
                    "#E41A1C", "#377EB8", "#7F0000",
                    "#35978f", "#FC8D62", "#2166ac",
                    "#E78AC3", "#A6D854", "#FFD92F",
                    "#E5C494", "#8DD3C7", "#6E016B" ,
                    "#BEBADA", "#e08214", "#80B1D3",
                    "#d6604d", "#ffff99", "#FCCDE5",
                    "#FF6A5A", "#BC80BD", "#CCEBC5" ,
                    "#fb9a99", "#B6646A", "#9F994E", 
                    "#7570B3" , "#c51b7d" ,"#66A61E" ,
                    "#E6AB02" , "#003c30", "#666666")
    
    mutationTypes <- stats::na.omit(unique(maf_data$Mutation_Type))
    
    # filter mutation types
    filteredTypes <- c()
    for (i in seq_len(length(mutationTypes))){
      if (length(grep(mutationTypes[i], mat)) != 0) {
        filteredTypes <- c(filteredTypes, mutationTypes[i])
      }
    }
    mutationTypes <- filteredTypes
    
    
    # sort types in legend
    if (class == "SP" | class == "SPCS") {
      sortType <- function(types) {
        publicType <- sort(types[grep("Public", types)])
        sharedType <- sort(types[grep("Shared", types)])
        privateType <- sort(types[grep("Private", types)])
        return(c(publicType, sharedType, privateType))
      }
      
      mutationTypes <- sortType(mutationTypes)
    } else {
      mutationTypes <- sort(mutationTypes)
    }
    
    col_type <- function(class) {
      # set.seed(123)
      cols <- colorScale[seq_len(length(mutationTypes))]
      names(cols) <- mutationTypes
      return(cols)
      
    }
    
    # prepare functions for assignment
    alter_fun_functions <- list()
    colorSelect <- col_type(class)
    for (type_num in seq_len(length(mutationTypes))){
      alter_fun_function <- paste0("function(x, y, w, h) grid::grid.rect(x, y, w * 0.9, h * 0.9,
                              gp = grid::gpar(fill = colorSelect[\'",mutationTypes[type_num], "\'],col = NA))")
      alter_fun_functions <- c(alter_fun_functions, eval(parse(text = alter_fun_function)))
    }
    names(alter_fun_functions) <- mutationTypes
    
    alter_fun <- function(class){
      l <- c(alter_fun_functions, Multi_hits = function(x, y, w, h)
        grid::grid.points(x, y, pch = 16, size = grid::unit(0.5, "char") 
        ), background = function(x, y, w, h)
          grid::grid.rect(x, y, w * 0.9, h * 0.9,
                          gp = grid::gpar(fill = bgCol, col = NA)))
      return(l)            
    }
  }
  # prepare legends
  
  ## type legend
  
  heatmapLegend <- ComplexHeatmap::Legend(title = "Type", 
                                          title_gp = grid::gpar(fontsize = legend.title.size),
                                          #title_gp = grid::gpar(fontsize = 11, fontface = "bold"),
                                          at = names(colorSelect),
                                          labels = gsub("_", "-", names(colorSelect)),
                                          labels_gp = grid::gpar(fontsize = legend.text.size),
                                          grid_width = unit(4, "mm"),
                                          grid_height = unit(4, "mm"), legend_gp = grid::gpar(fill = colorSelect))
  
  ## patient legend
  #patient.id <- unique(patient.split)
  
  if (is.null(patient.split)) {
    patient.id <- NULL
  } else if (removeEmptyCols ){
    excluded_sample_index <- c()
    for (i in seq_len(ncol(mat))) {
      if (all(is.na(mat[,i]))) {
        excluded_sample_index <- c(excluded_sample_index, i)
      }
    }
    
    if(!(is.null(excluded_sample_index))) {
      included_patients <- colnames(mat)[-excluded_sample_index]
    } else {
      included_patients <- colnames(mat)
    }
    patientID <- c()
    for (i in included_patients) {
      patientID <- c(patientID, strsplit(i, ":")[[1]][1])
    }
    patient.id <- unique(patientID)
  } else {
    included_patients <- colnames(mat)
    patient.id <- c()
    for (i in included_patients) {
      patient.id <- unique(c(patient.id, strsplit(i, ":")[[1]][1]))
    }
  }
  
  
  
  ## multi-hits legend
  multiLegend <- ComplexHeatmap::Legend(
    labels = "Multi_hits",
    labels_gp = grid::gpar(fontsize = legend.text.size),
    type = "points",
    pch = 16,
    grid_width = unit(4, "mm"),
    grid_height = unit(4, "mm")
  )
  
  if (is.null(patient.id)) {
    
    ## type-multi legend
    hm <- ComplexHeatmap::packLegend(heatmapLegend, multiLegend, direction = "vertical", gap = unit(0.3, "mm"))
    
    ## type-multi-patient legend
    hmp <- ComplexHeatmap::packLegend(hm, direction = "vertical", gap = unit(0.3, "cm"))
    
    ## type-patient legend
    hp <- ComplexHeatmap::packLegend(heatmapLegend, direction = "vertical", gap = unit(1.2, "cm"))
    
  }else {
    
    # set.seed(1234)
    if (is.null(patientsCol)) {
      qual_col_pals <- brewer.pal.info[brewer.pal.info$category == 'qual',]
      col_vector <- unlist(mapply(brewer.pal, qual_col_pals$maxcolors, rownames(qual_col_pals)))
      
      patientsCol <- col_vector[seq_len(length(patient.id))]
      names(patientsCol) <- patient.id
    } else {
      if (length(patientsCol) == length(patient.id)) {
        names(patientsCol) <- patient.id
      } else {
        stop("The number of provided colors does not equal to number of patients.")
      }
    }
    
    patientLegend <-  ComplexHeatmap::Legend(
      labels = patient.id, 
      legend_gp = grid::gpar(fill = patientsCol), 
      title_gp = grid::gpar(fontsize = legend.title.size),
      #title_gp = grid::gpar(fontsize = 11, fontface = "bold"),
      labels_gp = grid::gpar(fontsize = legend.text.size),
      grid_width = unit(4, "mm"),
      grid_height = unit(4, "mm"), title = "Patient")
    
    
    
    ## type-multi legend
    hm <- ComplexHeatmap::packLegend(heatmapLegend, multiLegend, direction = "vertical", gap = unit(0.3, "mm"))
    
    ## type-multi-patient legend
    hmp <- ComplexHeatmap::packLegend(hm, patientLegend, direction = "vertical", gap = unit(0.3, "cm"))
    
    ## type-patient legend
    hp <- ComplexHeatmap::packLegend(heatmapLegend, patientLegend, direction = "vertical", gap = unit(1.2, "cm"))
    
  }
  
  if (is.null(patient.split)) {
    
    ht <- suppressMessages(
      ComplexHeatmap::oncoPrint(
        mat,
        alter_fun = alter_fun(class),
        col = colorSelect,
        #column_title = "Mutational profile",
        column_title_gp = grid::gpar(fontsize = 13.5, col = "black"),
        #column_title_gp = grid::gpar(fontsize = 13.5, fontface = "bold", col = "black"),
        row_title_gp = grid::gpar(fontsize = 11, fontface = "plain", col = "black"),
        #heatmap_legend_param = heatmap_legend(class),
        show_heatmap_legend = FALSE,
        remove_empty_columns = removeEmptyCols,
        remove_empty_rows = removeEmptyRows,
        row_order = rowOrder,
        row_names_gp = grid::gpar(fontsize = gene.text.size, fontface = "italic", col = "black"),
        column_names_gp = grid::gpar(fontsize = sample.text.size, fontface = "plain", col = "black"),
        pct_digits = 2,
        pct_side = "right",
        row_names_side = "left", 
        #column_split = factor(patient.split,levels = unique(patient.split)),
        column_order = colnames(mat),
        column_labels = col_labels,
        show_column_names = showColnames,
        bottom_annotation = if(
          is.null(patient.split)) NULL else{
            ComplexHeatmap::HeatmapAnnotation(
              #df = data.frame(patient = colnames(mat)),
              df = data.frame(Patient = patient.split),
              show_annotation_name = FALSE,
              col = list(Patient = patientsCol),
              simple_anno_size = unit(0.2, "cm"),
              show_legend = FALSE,
              annotation_legend_param = list(
                title_gp = grid::gpar(fontsize = legend.title.size),
                labels_gp = grid::gpar(fontsize = legend.text.size),
                grid_width = unit(3.5, "mm"),
                grid_height = unit(3.5, "mm")
                #plot = FALSE
              )
              
            )}                
      )
    )
    
  } else {
    ht <- suppressMessages(
      ComplexHeatmap::oncoPrint(
        mat,
        alter_fun = alter_fun(class),
        col = colorSelect,
        column_title = NULL,
        column_title_gp = grid::gpar(fontsize = 13.5, col = "black"),
        #column_title_gp = grid::gpar(fontsize = 13.5, fontface = "bold", col = "black"),
        row_title_gp = grid::gpar(fontsize = 11, fontface = "plain", col = "black"),
        #heatmap_legend_param = heatmap_legend(class),
        show_heatmap_legend = FALSE,
        remove_empty_columns = removeEmptyCols,
        remove_empty_rows = removeEmptyRows,
        row_order = rowOrder,
        row_names_gp = grid::gpar(fontsize = gene.text.size, fontface = "italic", col = "black"),
        column_names_gp = grid::gpar(fontsize = sample.text.size, fontface = "plain", col = "black"),
        pct_digits = 2,
        pct_side = "right",
        row_names_side = "left", 
        column_split = factor(patient.split,levels = unique(patient.split)),
        column_order = colnames(mat),
        column_labels = col_labels,
        show_column_names = showColnames,
        bottom_annotation = if(
          is.null(patient.split)) NULL else{
            ComplexHeatmap::HeatmapAnnotation(
              #df = data.frame(patient = colnames(mat)),
              df = data.frame(Patient = patient.split),
              show_annotation_name = FALSE,
              col = list(Patient = patientsCol),
              simple_anno_size = unit(0.2, "cm"),
              show_legend = FALSE,
              annotation_legend_param = list(
                #title_gp = grid::gpar(fontsize = 10, fontface = "bold"),
                title_gp = grid::gpar(fontsize = legend.title.size),
                labels_gp = grid::gpar(fontsize = legend.text.size),
                grid_width = unit(3.5, "mm"),
                grid_height = unit(3.5, "mm")
                #plot = FALSE
              )
              
            )}                
      )
    )
  }
  
  if (multi_hit_exist) {
    ComplexHeatmap::draw(ht, heatmap_legend_list = hmp,
                         padding = unit(c(3, 3, 3, 3), "mm"))
    
  } else {
    ComplexHeatmap::draw(ht, heatmap_legend_list = hp,
                         padding = unit(c(3, 3, 3, 3), "mm"))
  }
  
  
}