前言

Immugent在前两期推文中：CommPath（一）：鉴定特定细胞的互作关系和CommPath（二）：鉴定关键细胞互作信号介导的功能通路中，分别介绍了如何使用CommPath联合解析特定的信号通路和相关的功能通路。那么，在本期推文中，Immugent就来通过代码实操的方式介绍，如何使用CommPath将这两部分内容连接起来说故事。

想要复现本期推文代码的小伙伴需要跑一下前两个流程~~

代码流程

Identification of pathway-mediated cell-cell communication chain

# To investigate the activated patways and the associated receptors for a specific cluster
select.ident = 'Endothelial'
pdf('pathPlot.Endothelial.pdf',height=6,width=10)
pathPlot(object = tumor.obj, 
    select.ident = select.ident, 
    acti.path.dat = acti.path.filtered.dat)
dev.off()

image.png

# To select self-defind cluster for the  central cell population
# And select pathways of interest by names
# Also select the top N receptors by interaction intensity 
select.ident = 'Endothelial'
pdf('pathPlot.Endothelial.selectpathway.pdf',height=4,width=10)
pathPlot(object = tumor.obj, 
    select.ident = select.ident, 
    up.ident = c("Treg","Myeloid","Endothelial","Fibroblast","Hepatocyte"),
    select.path = c("JAK-STAT signaling pathway", "MAPK signaling pathway", "Ras signaling pathway", "PI3K-Akt signaling pathway", "AMPK signaling pathway"),
    top.n.receptor = 5,
    acti.path.dat = acti.path.filtered.dat)
dev.off()

image.png

# To investigate the activated patways, the associated receptors and ligands for a specific cluster
pdf('pathChainPlot.Endothelial.pdf',height=6,width=14)
pathChainPlot(object = tumor.obj, 
    select.ident = select.ident, 
    up.ident = c("Treg","Myeloid","Endothelial","Fibroblast","Hepatocyte"),
    down.ident = c("B","Myeloid","Endothelial","Fibroblast","Hepatocyte"),
    select.path = c("JAK-STAT signaling pathway", "MAPK signaling pathway", "Ras signaling pathway", "PI3K-Akt signaling pathway", "AMPK signaling pathway"),
    acti.path.dat = acti.path.filtered.dat)
dev.off()

image.png

Comparison of cell-cell communication between two conditions

# load(url("https://figshare.com/ndownloader/files/35185525"))
load("path_to_download/HCC.normal.3k.RData")

select.ident <- 'Endothelial'
diff.path.dat <- comparePath(object.1 = tumor.obj, 
  object.2 = normal.obj, 
  select.ident = select.ident)

select.ident <- 'Endothelial'
diff.path.dat <- comparePath(object.1 = tumor.obj, 
  object.2 = normal.obj, 
  select.ident = select.ident)
    
# To compare differentially activated pathways and the involved receptors between the selected clusters in 2 CommPath objects
pdf('pathPlot.compare.Endothelial.pdf',height=6,width=10)
pathPlot.compare(object.1 = tumor.obj, 
  object.2 = normal.obj, 
  select.ident = select.ident,
  up.ident = c("Treg","Myeloid","Endothelial","Fibroblast","Hepatocyte"),
  diff.path.dat = diff.path.dat)
dev.off()

image.png

# To compare the pathway-mediated cell-cell communication chain for a specific cluster between 2 CommPath objects
pdf('pathChainPlot.compare.Endothelial.pdf',height=6,width=14)
pathChainPlot.compare(object.1 = tumor.obj, 
  object.2 = normal.obj, 
  select.ident = select.ident, 
  up.ident = c("Myeloid","Endothelial","Fibroblast","Hepatocyte"),
    down.ident = c("CD8 T","CD4 T","Treg","NK","Endothelial","Fibroblast","Hepatocyte"),
    select.path = c("Gastric cancer", "Relaxin signaling pathway", "Axon guidance"),
  diff.path.dat = diff.path.dat)
dev.off()

image.png

说在最后

从上面的分析结果我们可以看出，CommPath包可以很好的将目标细胞之间的信号通讯和潜在相关的功能通路联系起来，而且还给出了具体的信号分子。不过有一说一，上面示例中的互作信号和下游通路之间的关系并没有很好的对应关系，可能是因为输入的信号通路不太适合相应的生物学场景。因此，想得到相对可靠的生信分析结果，除了需要具备较好的工具条件，还需要找到很好的适用生物学场景。好啦，截止到目前，有关CommPath包的所有推文就都更新完毕了，感兴趣的小伙伴赶紧用起来吧。

本期分享到这里就结束了，我们下期再会~~