Chapter 9 Adding gene labels

Here shows an example to add gene names from gtf file on genome track. You only need to extract the gene regions which inlude seqname, start,end and geneName. That’s enough.

---

Process gtf file:

library(rtracklayer)
library(tidyverse)
library(ggcirclize)

gtf <- import.gff("Homo_sapiens.GRCh38.110.chr.gtf.gz") %>% as.data.frame()

sample_genes <- sample(unique(gtf$gene_name),40,replace = F)

genes_region <- gtf %>% filter(gene_name %in% sample_genes & type == "gene") %>% 
  mutate(seqnames = paste0("chr",seqnames))

# check
head(genes_region[,1:7],3)
#   seqnames     start       end width strand         source type
# 1     chr1  17066761  17119451 52691      - ensembl_havana gene
# 2     chr1 111490317 111490423   107      -        ensembl gene
# 3     chr1 161623196 161631963  8768      - ensembl_havana gene

sample_genes2 <- sample(unique(gtf$gene_name),40,replace = F)

genes_region2 <- gtf %>% filter(gene_name %in% sample_genes2 & type == "gene") %>% 
  mutate(seqnames = paste0("chr",seqnames))

Plot:

data("hg38_chrom_info")
cytoband_hg38 <- hg38_chrom_info$cytoband

ggcirclize(cytoband_hg38,
           aes(end = 360,genome = "hg38",
               chr = chr,gstart = start,gend = end)) +
  ggcirclize::geom_trackgenomicrect(aes(r0 = 0.8,r1 = 0.85,fill = stain),
                                    color = NA,add.xaxis = F,strip.label = F) +
  scale_fill_manual(values = c("gneg" = "white","gpos25" = "grey75","gpos50" = "grey50",
                               "gpos100" = "black","gvar" = "black","acen" = "red",
                               "stalk" = "blue")) +
  ggcirclize::geom_trackgenomiclabel2(data = genes_region,
                                      aes(r0 = 0.7,r1 = 0.7,label = gene_name,
                                          chr = seqnames,gstart = start,gend = end,
                                          color = seqnames),
                                      keep.all.chrom = T,
                                      strip.label = F) +
  ggcirclize::geom_trackgenomiclabel(data = genes_region2,
                                      aes(r0 = 0.95,r1 = 0.95,label = gene_name,
                                          chr = seqnames,gstart = start,gend = end),
                                     link_pos = "bottom",
                                      keep.all.chrom = T,
                                      strip.label = F)