Skip to contents

Visualize Gene-Gene Correlation in TCGA

Usage

vis_pcawg_gene_cor(
  Gene1 = "CSF1R",
  Gene2 = "JAK3",
  data_type1 = "mRNA",
  data_type2 = "mRNA",
  cor_method = "spearman",
  purity_adj = TRUE,
  use_log_x = FALSE,
  use_log_y = FALSE,
  use_regline = TRUE,
  dcc_project_code_choose = "BLCA-US",
  use_all = FALSE,
  filter_tumor = TRUE,
  alpha = 0.5,
  color = "#000000",
  opt_pancan = .opt_pancan
)

Arguments

Gene1

a molecular identifier (e.g., "TP53") or a formula specifying genomic signature ("TP53 + 2 * KRAS - 1.3 * PTEN").

Gene2

a molecular identifier (e.g., "TP53") or a formula specifying genomic signature ("TP53 + 2 * KRAS - 1.3 * PTEN").

data_type1

choose gene profile type for the first gene, including "mRNA","transcript","methylation","miRNA","protein","cnv_gistic2"

data_type2

choose gene profile type for the second gene, including "mRNA","transcript","methylation","miRNA","protein","cnv_gistic2"

cor_method

correlation method

purity_adj

whether performing partial correlation adjusted by purity

use_log_x

if TRUE, log X values.

use_log_y

if TRUE, log Y values.

use_regline

if TRUE, add regression line.

dcc_project_code_choose

select project code.

use_all

use all sample, default FALSE.

filter_tumor

whether use tumor sample only, default TRUE

alpha

dot alpha.

color

dot color.

opt_pancan

specify one dataset for some molercular profiles

Value

a ggplot object