RangedSummarizedExperiment类是一个类似矩阵的容器,其中行表示感兴趣的范围(作为GRanges或GRangesList对象),列表示样本(样本数据汇总为DataFrame)。RangedSummarizedExperiment包含一个或多个分析,每个分析由数值或其他模式的类似矩阵的对象表示。
RangedSummarizedExperiment是SummarizedExperiment的一个子类,因此,SummarizedExperiment类中记录的所有方法也适用于RangedSummarizedExperiment对象。
Usage
SummarizedExperiment(assays=SimpleList(), rowData=NULL, rowRanges=GRangesList(), colData=DataFrame(), metadata=list())
1. 生成SummarizedExperiment对象
library(SummarizedExperiment)
# construction
nrows <- 200
ncols <- 6
counts <- matrix(runif(nrows * ncols, 1, 1e4), nrows)
rowRanges <- GRanges(rep(c("chr1", "chr2"), c(50, 150)),
IRanges(floor(runif(200, 1e5, 1e6)), width=100),
strand=sample(c("+", "-"), 200, TRUE),
feature_id=sprintf("ID%03d", 1:200))
colData <- DataFrame(Treatment=rep(c("ChIP", "Input"), 3),
row.names=LETTERS[1:6])
SummarizedExperiment(assays=list(counts=counts),
rowRanges=rowRanges, colData=colData)
2. 查看行,列以及元数据
library(SummarizedExperiment)
data(airway, package="airway")
se <- airway
se
# 行(特征)信息
rowRanges(se) # rowData() for SummarizedExperiment class
rownames(se)
# 列(样本)信息
colData(se)
colnames(colData(se))
# 元数据
metadata(se) #list
# add to list
metadata(se)$formula <- counts ~ dex + albut
metadata(se)
3. 取子集
对SummarizedExperiment或RangedSummarizedExperiment对象取子集,得到的还是SummarizedExperiment或RangedSummarizedExperiment对象。
# subset
se[, se$dex == "trt"]
# subset the first five transcripts and first three samples
se[1:5, 1:3]
se[, se$cell == "N61311"]
se["ENSG00000000003", se$cell == "N61311"]
# Subset for only rows which are in the interval 100,000 to 110,000 of
# chromosome 1 选择特定位置的features
roi <- GRanges(seqnames="1", ranges=100000:1100000)
subsetByOverlaps(se, roi)
counts <- matrix(1:15, 5, 3, dimnames=list(LETTERS[1:5], LETTERS[1:3]))
dates <- SummarizedExperiment(assays=list(counts=counts),
rowData=DataFrame(month=month.name[1:5], day=1:5))
# Subset all January assays
dates[rowData(dates)$month == "January", ]
4. 得到基因表达矩阵
# assays() operates on the entire list of assay data as a whole,
# while assay() operates on only one assay at a time.
assays(se)$counts
assays(se)[[1]][1:5, 1:5]
# assay defaults to the first assay if no i is given
assay(se)[1:5, 1:5]
assay(se, 1)[1:5, 1:5]
class(assays(se)) # "SimpleList"
class(assay(se)) #"matrix" "array"
注意:assay() 和assays()的区别。