Error:The design matrix has the same number of samples and coefficients to fit, so estimation of dispersion is not possible. Treating samples as replicates was deprecated in v1.20 and no longer su
2
0
Entering edit mode
takisliako • 0
@takisliako-24300
Last seen 22 months ago

Hi everyone,

I've been trying to use DESeq2 to look at expression differences across 4 SRR: 2967085, 2967086, 2973633 and 23351735.

I've used HTseq and got 4 files. An example of the files:

uc001aaa.3  0
uc001aac.4  0
uc001aae.4  0
uc001aah.4  0
uc001aai.1  0
uc001aak.3  0
uc001aal.1  0
uc001aaq.2  0
uc001aar.2  0
uc001aau.3  2

Then i tried:

sampleFiles <- grep("*S",list.files(directory),value=TRUE)
sampleCondition <- sub("(.*S).*","\\1",sampleFiles)
sampleTable <- data.frame(sampleName = sampleFiles,
                          fileName = sampleFiles,
                          condition = sampleCondition)
sampleTable$condition <- factor(sampleTable$condition)
sampleCondition
[1] "AS" "BS" "CS" "DS"

library("DESeq2")
ddsHTSeq <- DESeqDataSetFromHTSeqCount(sampleTable = sampleTable,
                                       directory = directory,
                                       design= ~ condition)
ddsHTSeq
dim: 82960 4 
metadata(1): version
assays(1): counts
rownames(82960): uc001aaa.3 uc001aac.4 ... uc031tky.1 uc031tkz.1
rowData names(0):
colnames(4): AS22967085.txt BS22967086.txt CS22973633.txt
  DS23351735.txt
colData names(1): condition

And then when i try:

dds <- DESeq(ddsHTSeq)
estimating size factors
estimating dispersions
Error in checkForExperimentalReplicates(object, modelMatrix) : 

  The design matrix has the same number of samples and coefficients to fit,
  so estimation of dispersion is not possible. Treating samples
  as replicates was deprecated in v1.20 and no longer supported since v1.22.

I'm new in learning DESeq2. Can anyone help?

deseq2 • 5.6k views
ADD COMMENT
0
Entering edit mode
@james-w-macdonald-5106
Last seen 15 hours ago
United States

The error is telling you that you only have one sample for each sample type. In that situation you can't do any statistical analysis, because you need replicates to estimate the biological variability. It's possible that you have two groups with two samples in each, in which case you need to change your 'condition' to reflect that.

ADD COMMENT
0
Entering edit mode

Dear James,

I also have same issue. conditions. In my case conditions are normal and cancer and file name mentioned as N and C. Kindly tell. How to change and where to change

ADD REPLY
0
Entering edit mode

In general you should create a new post rather than hopping on old posts. Also, please read the FAQ so you know how to ask a good question. As it stands your question is unanswerable.

ADD REPLY
0
Entering edit mode

You cannot compare two samples with DESeq2.

ADD REPLY
0
Entering edit mode
sunnet.ys • 0
@8e7a88b3
Last seen 4 months ago
Germany

Hi Neha, and of course anyone who may encounter this kind of issue in the future. Sometimes you actually have count data that is duplicate or more and you are wondering "but my data is duplicate, why am i still getting this error", then in this case i advice you check you colData table, choose a column of design parameter that reflects your true experimental conditions and use it for your design in the deseq2 functional analysis. It will solve the error.

ADD COMMENT

Login before adding your answer.

Traffic: 199 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