Question: How to get normalized counts and do differential expression using time series data (gene knock down) using DESeq2
0
11 months ago by
dryellaboina0 wrote:

I have 3 time points (0h, 120h, 240h) data in replicates. I need to find out the effect of a gene knock-down on cell death mechanisms i.e., genes affected by knock-down. Is 3 times points are enough to consider as time series data. Shall I do pairwise comparision of different time points or do time series analysis sing DESeq2. Here is an example code. Please help me with a code to get normalzed counts as well. Here is an example code I have written after going through DESeq2 Manual.

library(DESeq2) Time = rep(c("0h", "120h", "240h"), each = 2)

Treat = c(rep("Control", 2), rep("Treat", 4))

nameD <- paste(Treat, Time, c(rep(LETTERS[1:2], 1), rep(LETTERS[3:4], 2)), sep = "_")

sampleInfo <- data.frame(row.names = nameD, Time = Time, Treat = Treat)

colnames(countdata)=row.names(sampleInfo)

ddsMat <- DESeqDataSetFromMatrix(countData=countdata, colData=sampleInfo, design=~ Time + Treat+Time:Treat)

I am getting following error with above code

Error in checkFullRank(modelMatrix) : the model matrix is not full rank, so the model cannot be fit as specified. One or more variables or interaction terms in the design formula are linear combinations of the others and must be removed.

deseq2 • 149 views
modified 11 months ago by Michael Love23k • written 11 months ago by dryellaboina0
Answer: How to get normalized counts and do differential expression using time series da
0
11 months ago by
Michael Love23k
United States
Michael Love23k wrote:

Control and time 0hr are identical (there are no control samples after time 0, and no treated samples at time 0). So you should just drop the "treat" variable from the design. All of the information about the experiment is contained in the "time" variable.