见上篇:genome browser | ggplot | 基因组可视化 | R | transcript | isoform
新需求(画出Fig 4D上面的辅图, Science, 2021):
山寨版,分析解构:
- 首先去method里找,完全没写是什么工具画的,那就猜吧;
- 其次看配色和对齐方式,基本肯定是ggplot画的,但仍然查不到是什么工具;
- 最后,能猜到这是卡通图,并不是真实的数据(exon的长度过于平均);
- 优点:辅图一目了然的AS event可视化,这种barplot的可视化(颜色对应,PSI和为1)更加易读,可以看出作者在AS可视化上的功底;
- 细节也很到位:箭头和UTR
做一个API,自动读取AS event格式,添加箭头和UTR,以后可以常用。
不难
举个栗子:
# prepare kartoon data # SE, Cassettes AS_1 <- data.frame(EXONSTART=c(0,2,4), EXONEND=c(1,3,5), EXONSTRAND="+") seg_1 <- data.frame(x=c(0, 1.5, 3.5), xend=c(1.5, 3.5, 4.5), y=0, yend=0) AS_2 <- data.frame(EXONSTART=c(0,4), EXONEND=c(1,5), EXONSTRAND="+") seg_2 <- data.frame(x=c(0, 1.5, 2.5, 3.5), xend=c(1.5, 2.5, 3.5, 4.5), y=0, yend=0) options(repr.plot.width=2.5, repr.plot.height=1) p1 <- ggplot() + # add line and arrow # geom_hline(yintercept=0)+ # geom_segment(x = 0, xend = 5.5, y = 0, yend = 0) + geom_segment(data=seg_1, aes(x=x,y=y,xend=xend,yend=yend), arrow=arrow(length=unit(0.3,"cm")), size=1) + # add exons geom_rect(data=AS_1, aes(xmin=EXONSTART, xmax=EXONEND,ymin= -0.1,ymax=0.1),fill="#4DAF4A")+ # add UTR # geom_rect(data=UTR, aes(xmin=EXONSTART, xmax=EXONEND,ymin= -0.05,ymax=0.05),fill="#282a73")+ # details labs(title = NULL,subtitle = NULL)+ theme_void() + theme(legend.position = "none") + scale_x_continuous(expand = c(0.1, 0.1), limits = c(0, 5)) + scale_y_continuous(expand = c(0.02, 0.02), limits = c(-0.1, 0.1)) p1 options(repr.plot.width=2.5, repr.plot.height=1) p2 <- ggplot() + # add line and arrow # geom_hline(yintercept=0)+ # geom_segment(x = 0, xend = 5.5, y = 0, yend = 0) + geom_segment(data=seg_2, aes(x=x,y=y,xend=xend,yend=yend), arrow=arrow(length=unit(0.3,"cm")), size=1) + # add exons geom_rect(data=AS_2, aes(xmin=EXONSTART, xmax=EXONEND,ymin= -0.1,ymax=0.1),fill="#984EA3")+ # add UTR # geom_rect(data=UTR, aes(xmin=EXONSTART, xmax=EXONEND,ymin= -0.05,ymax=0.05),fill="#282a73")+ # details labs(title = NULL,subtitle = NULL)+ theme_void() + theme(legend.position = "none") + scale_x_continuous(expand = c(0.1, 0.1), limits = c(0, 5))+ scale_y_continuous(expand = c(0.02, 0.02), limits = c(-0.1, 0.1)) p2 options(repr.plot.width=2.5, repr.plot.height=2) AS1 <- cowplot::plot_grid(p1,p2,ncol = 1) AS1 options(repr.plot.width=9, repr.plot.height=1) cowplot::plot_grid(AS2, AS2, AS3, AS1,ncol = 4)
最终的模仿图,颜值还行:
分析案例:project/scPipeline/AS/suppa_AS_stat.ipynb
参考: