how to normalize data for rna-seq after limma for heatmap or cox
1
1
Entering edit mode
linouhao ▴ 20
@linouhao-15901
Last seen 20 months ago
United States

how to normalize data for rna-seq after limma for heatmap or cox or other downstream analysis, should I use the same way as in edger, jusr use the logCPM data here thanks a lot

dge <- DGEList(counts=mRNAdata)
dge <- calcNormFactors(dge)
**logCPM** <- cpm(dge, log=TRUE, prior.count=3)

v <- voom(dge,design,plot=TRUE, normalize="quantile")
fit <- lmFit(v, design)
cont.matrix=makeContrasts(contrasts=c('tumor-normal'),levels = design)
fit2=contrasts.fit(fit,cont.matrix)
fit2=eBayes(fit2)
tempOutput = topTable(fit2, coef='tumor-normal', n=Inf)
DEG_limma_voom = na.omit(tempOutput)
limma • 2.5k views
ADD COMMENT
1
Entering edit mode
Kevin Blighe ★ 4.0k
@kevin
Last seen 6 weeks ago
Republic of Ireland

Generally, yes, you should use the expression values contained in your logCPM variable for downstream clustering (heatmap) and Cox regression. For the heatmap, you may additionally want to center and/or scale the data (to Z-scores) in order to 'improve' visualisation in relation to the heatmap colours. For Cox regression, in a survival analysis context, you may want to dichotomise the variable by median (but there are limitations to this), or categorise it into deciles, quartiles, or something else in order to generate strata for easier interpretatation.

Kevin

ADD COMMENT
0
Entering edit mode

so limma does not have its own method, but can only use the function of edger? can I use the result of voom?

ADD REPLY
1
Entering edit mode

You mean functions for heatmaps and Cox regression? - for Heatmaps, you can use gplots, pheatmap, or ComplexHeatmap ( see here: https://github.com/kevinblighe/E-MTAB-6141 ). For Cox regression, you can use survival::coxph

For the heatmap, in a typical scenario, you would subset your logCPM data-matrix for statistically significant genes identified via topTable(), and then perform clustering on this subset data-matrix.

ADD REPLY
0
Entering edit mode

no, I am not meaning this I am talking about whether I can use the voom data for downtream analysis instead of logcpm in packages of edger.

ADD REPLY

Login before adding your answer.

Traffic: 386 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6