Why less DE for non replicates

Question

[manoharankumar01]

Hi,

I have been working with human and mouse organisms with non replicates, I am getting less number of DE genes using different method DESeq. Below is the total read count for two samples.

human: sample1: 22905420 sample2: 23715800 total genes: 25,370 DE genes: 97

I am trying to understand why did i get less DE genes? Could someone help me?

Thanks.

[Nicolas Rosewick]

As always without any replicates it would be hard to draw conclusions.. IMO I would not trust the results..

[Devon Ryan]

We need to post this in a banner at the top of the page.

[WouterDeCoster]

Or just a helpful pop-up when making a post containing some keywords: https://cdn.meme.am/cache/instances/folder314/500x/73204314.jpg

[WouterDeCoster]

Please be more informative with regard to the experimental setup and the commands you used, don't make us guess.

[manoharankumar01]

Sorry, I have used following commands:

library(DESeq)

data = read.table("DEinput.txt", header=T, row.names=1, com='') col_ordering = c(1,2) rnaseqMatrix = data[,col_ordering] rnaseqMatrix = round(rnaseqMatrix) rnaseqMatrix = rnaseqMatrix[rowSums(rnaseqMatrix)>=2,] conditions = factor(c(rep("s1", 1), rep("s2", 1)))

exp_study = newCountDataSet(rnaseqMatrix, conditions) exp_study = estimateSizeFactors(exp_study) exp_study = estimateDispersions(exp_study, method="blind", sharingMode="fit-only")

res = nbinomTest(exp_study, "s1", "s2")

[poisonAlien]

You don't have replicates. Hence, test is limited in degrees of freedom to estimate variance and thereby less power to detect DE genes.

Answer 1

[Devon Ryan]

Fewer replicates means less power.