Pretty stacked oncoplot

Pretty stacked oncoplot

Usage

prettyStackedOncoplot(
  these_samples_metadata,
  maf_data,
  metadataColumns = "pathology",
  sortByMetadataColumns,
  seg_data,
  sortByPGA = FALSE,
  cn_state_matrix = NULL,
  ashm_matrix,
  regions_bed,
  genes,
  sortByGenes,
  genes_CN_thresh,
  secondPlotType = "pretty_CN_heatmap",
  oncoplot_location = "top",
  cluster_samples = FALSE,
  secondPlotArgs,
  oncoplotArgs,
  returnEverything = FALSE,
  plot_width = 15,
  oncoplotHeight = 6,
  secondPlotHeight = 6,
  verbose = FALSE,
  row_names_side = "right",
  pctFontSize = 0
)

Arguments

these_samples_metadata

A metadata data frame with sample_id as a column. The order of sample IDs in the rows of this data frame will dictate the order of samples in the oncoplot.

maf_data

A data frame with maf data that will be used to populate the oncoplot. Required parameter.

metadataColumns

A character vector of column names that specifies which columns from these_samples_metadata will be displayed below the oncoplot.

secondPlotType

Defaults to pretty_CN_heatmap, which is currently the only option tested with this function.

verbose

Set to TRUE for ultra chatty mode

Examples

suppressMessages(
  suppressWarnings({

library(GAMBLR.open)

# Prepare some metadata
 dlbcl_genome_meta = get_gambl_metadata() %>% 
  dplyr::filter(pathology=="DLBCL",
                seq_type=="genome") %>%
  check_and_clean_metadata(.,duplicate_action="keep_first")

# Get CN segments for these samples
dlbcl_seg = get_cn_segments(dlbcl_genome_meta)

# Prepare CN matrix
cn_mat = segmented_data_to_cn_matrix(dlbcl_seg,
                                     these = dlbcl_genome_meta,
                                     adjust_for_ploidy = TRUE)

dlbcl_maf = get_all_coding_ssm(dlbcl_genome_meta)

}))

genes=c("KMT2D","BCL2","CREBBP","EZH2","MYD88","CD79B","TP53",
        "PIM1","CARD11","SGK1","SOCS1",'TET2',"SPEN",
        "ETV6","CD83","B2M","S1PR2","GNA13","BTG1",
        "BTG2","DDX3X","KLHL6","HIST1H1E","TBL1XR1","SMARCA4")

# we will only sort on the mutation status of these
# genes and in this order
sortGenes = c("TP53","KMT2D","BCL2","EZH2","MYD88","CD79B")


 CN_thresh = c("REL"=4,
              "CDKN2A"=1,
              "MIR17HG"=4,
              "TP53"=1,
              "TNFRSF14"=1,
              "TNFAIP3"=1)

# oncoplot on top with column order dicted by the mutation status
# of sortGenes (in the order of the genes appearance in that vector)

suppressMessages(
  suppressWarnings({
prettyStackedOncoplot(these_samples_metadata = dlbcl_genome_meta,  
                        maf_data = dlbcl_maf,
                        metadataColumns = c("pathology","lymphgen"),
                        sortByMetadataColumns = c("pathology","lymphgen"),
                        cn_state_matrix = cn_mat,
                        genes_CN_thresh = CN_thresh,
                        genes = genes,
                        sortByGenes = sortGenes)

}))

#> NULL
# oncoplot on top. Clustering of mutations is used to order the columns.
suppressMessages(
  suppressWarnings({
prettyStackedOncoplot(these_samples_metadata = dlbcl_genome_meta,  
                      maf_data = dlbcl_maf,
                      metadataColumns = c("pathology","lymphgen"),
                      cluster_samples = TRUE,
                      cn_state_matrix = cn_mat,
                      genes_CN_thresh = CN_thresh,
                      genes = genes)

}))

#> NULL
# make a list of arguments for the second (here, the upper) plot
CN_args = list("keep_these_chromosomes"=c("2"),
              "scale_by_sample" = TRUE)
suppressMessages(
  suppressWarnings({

prettyStackedOncoplot(these_samples_metadata = dlbcl_genome_meta,
                     maf_data = dlbcl_maf,
                     sortByGenes = "REL",
                     metadataColumns = c("pathology","lymphgen"),
                     oncoplot_location = "bottom",
                     secondPlotArgs = CN_args,
                     cn_state_matrix = cn_mat,
                     genes_CN_thresh = CN_thresh,
                     genes = genes,
                     oncoplotHeight = 8,
                     secondPlotHeight=3)

}))
#> [1] "REL_cn"
#> [1] "sample_id"   "TNFRSF14_cn" "REL_cn"      "TNFAIP3_cn"  "CDKN2A_cn"  
#> [6] "MIR17HG_cn"  "TP53_cn"    

# make a list of arguments for the second (here, the upper) plot
CN_args = list(
              "scale_by_sample" = TRUE,
              "hide_these_chromosomes" = "X")

# Specifying sortByGenes automatically ensures those genes
# are sorted by their CN status then mutation status when
# oncoplot_location is "bottom".
# The order in the upper plot restricts the order of the lower plot.

suppressMessages(
  suppressWarnings({
prettyStackedOncoplot(these_samples_metadata = dlbcl_genome_meta,
                     maf_data = dlbcl_maf,
                     sortByGenes = "TP53",
                     metadataColumns = c("pathology","lymphgen"),
                     oncoplot_location = "bottom",
                     secondPlotArgs = CN_args,
                     cn_state_matrix = cn_mat,
                     genes_CN_thresh = CN_thresh,
                     genes = genes,
                     secondPlotHeight=9)
}))
#> [1] "TP53_cn"
#> [1] "sample_id"   "TNFRSF14_cn" "REL_cn"      "TNFAIP3_cn"  "CDKN2A_cn"  
#> [6] "MIR17HG_cn"  "TP53_cn"    


some_regions = create_bed_data(GAMBLR.data::grch37_ashm_regions,
                              fix_names = "concat",
                              concat_cols = c("gene","region"),sep="-")

suppressMessages(
  suppressWarnings({

simple_ashm_mat <-
 get_ashm_count_matrix(some_regions,
                       these_samples_metadata = dlbcl_genome_meta)



prettyStackedOncoplot(these_samples_metadata = dlbcl_genome_meta,
                     maf_data = dlbcl_maf,
                     regions_bed= some_regions,
                     metadataColumns = c("pathology","lymphgen"),
                     oncoplot_location = "bottom",
                     ashm_matrix = simple_ashm_mat,
                    secondPlotType = "prettyMutationDensity",
                     secondPlotArgs = list("merge_genes"=TRUE,
                                           region_fontsize=3),
                     genes = genes,
                    cluster_samples = TRUE,
                    secondPlotHeight = 9)

}))