scRNA---Day6

cell cycle

细胞周期相关的包：https://mp.weixin.qq.com/s/nFSa5hXuKHrGu_othopbWQ
library(scran)
sce <- SingleCellExperiment(list(counts=dat))
cyclone()函数：

sce: SingleCellExperiment
mm.pairs(mouse的对象)
scran包自带mouse_cycle_markers.rds & human_cycle_markers.rds

library(org.Mm.eg.db)
mm.pairs <- readRDS(system.file("exdata", "mouse_cycle_markers.rds", 
                                package="scran"))

ensembl <- mapIds(org.Mm.eg.db, keys=rownames(sce), 
                  keytype="SYMBOL", column="ENSEMBL")

cyclone函数需要识别ensembl，利用org.Mm.eg.db包根据symbol添加ensembl

判断周期
assigned <- cyclone(sce, pairs=mm.pairs, gene.names=ensembl) save(assigned,file = 'cell_cycle_assigned.Rdata')

Rplot.png

rpkm

counts / （基因长度*文库大小）
基因长度包：TxDb.Mmusculus.UCSC.mm10.knownGene
txDb对象：？TxDb：the TxDb class is a container for storing transcript annotations.
exons()函数取exon
genes()函数取gene
findoverlaps()取交叉
?GRanges:The GRanges class is a container for the genomic locations and their associated annotations.
transcriptLengths()转录本长度函数
!duplicate() 删除重复即后出现的值
应用到的几个函数
cbind：根据列进行合并，即叠加所有列，m列的矩阵与n列的矩阵cbind()最后变成m+n列，合并前提：cbind(a, c)中矩阵a、c的行数必需相符
rbind：根据行进行合并，就是行的叠加，m行的矩阵与n行的矩阵rbind()最后变成m+n行，合并前提：rbind(a, c)中矩阵a、c的列数必需相符
split():split(x, f, drop = FALSE, ...) x data to be divided，f define the grouping
split(d$income, list(d$gender,d$over25)) 将income按照gender、over25分组