单细胞RNA测序揭示了人膀胱癌上皮细胞异质性和侵袭性亚群

#samples
GSM4006644 BC1-scRNA
GSM4006645 BC2-scRNA
GSM4006646 BC3-scRNA
GSM4006647 BC4-scRNA
GSM4006648 BC5-scRNA
GSM4751267 BC6-scRNA
GSM4751268 BC7-scRNA
GSM5329919 BCN-scRNA

#数据格式
GSM4006644_BC1_gene_cell_exprs_table.txt.gz 11.7 Mb
GSM4006645_BC2_gene_cell_exprs_table.txt.gz 13.1 Mb
GSM4006646_BC3_gene_cell_exprs_table.txt.gz 8.1 Mb
GSM4006647_BC4_gene_cell_exprs_table.txt.gz 8.1 Mb
GSM4006648_BC5_gene_cell_exprs_table.txt.gz 18.3 Mb
GSM4751267_BC6_gene_cell_exprs_table.txt.gz 14.3 Mb
GSM4751268_BC7_gene_cell_exprs_table.txt.gz 11.1 Mb
GSM5329919_BCN_gene_cell_exprs_table.xls.gz 11.4 Mb

dir='GSE135337_RAW/' 
samples=list.files( dir ,pattern = 'gz')
samples 
library(data.table)
ctList = lapply(samples,function(pro){ 
  # pro=samples[1] 
  print(pro)
  ct=fread(file.path( dir ,pro),data.table = F)
  ct[1:4,1:4]
  rownames(ct)=ct[,1]
  colnames(ct) = paste(gsub('.txt.gz','',pro),
                       colnames(ct) ,sep = '_')
  ct=ct[,-1] 
  return(ct)
})

#合并样品
lapply(ctList, dim)
tmp =table(unlist(lapply(ctList, rownames)))
cg = names(tmp)[tmp==length(samples)]
bigct = do.call(cbind,
                lapply(ctList,function(ct){ 
                  ct = ct[cg,] 
                  return(ct)
                }))
sce.all=CreateSeuratObject(counts =  bigct, 
                           min.cells = 5,
                           min.features = 300)
sce.all
as.data.frame(sce.all@assays$RNA$counts[1:10, 1:2])
head(sce.all@meta.data, 10)
table(sce.all@meta.data$orig.ident)