Open gmgitx opened 5 years ago
##################数据按DESeq2 document速览##################
##################数据按DESeq2要求给入##################
expression values
Note:
对于样本的描述
Expression Set Object包含a和b,并且给出design formula。 Note:
##################数据处理部分################## #########首先面对的是Normalization
需要讨论的: TMM等,与scale和log三者关系: TMM等,scale应该都可以称为normalize log: transformed
ExpressionNormalizationWorkflow
biostars上的详细讨论 PVCA 分为监督和非监督的
哈佛DGE简介,解释reduced和LRT,以及not log2 fold changes而是从加强P cutoff
reduced:for test="LRT", a reduced formula to compare against
collapseReplicates()解读 the term techical replicate implies multiple sequencing runs of the same library.
cds <- newCountDataSetFromHTSeqCount(sampleTable=sampleTable, directory=directory) cds <- estimateSizeFactors(cds) cds <- estimateDispersions(cds) data.frame(sizefactors=sizeFactors(cds), rawcounts=colSums(counts(cds, normalized=FALSE)))
DESeq()函数的理解,引用自青山屋主文章
DESeq包含三步,estimation of size factors(estimateSizeFactors), estimation of dispersion(estimateDispersons), Negative Binomial GLM fitting and Wald statistics(nbinomWaldTest),可以分布运行,也可用一步到位,最后返回 results可用的DESeqDataSet对象。
25