Matrix not full rank (time-series experiment with one group and covariates )
1
0
Entering edit mode
heikki.sarin ▴ 10
@heikkisarin-13379
Last seen 4.0 years ago

I have problems performing a time-series experiment with DESeq2 while accounting for age as a covariate. We have one group with each individual having three time points. We're interested in detecting differential effects between time points while accounting for within individual variability and age as a covariates. The following code runs just fine but once adding age into the equation DESeq2 gives an error that the model matrix is not full rank. I suspect the problem being with variables SUBJECT_ID and AGE being linear combination of each other based on the DESeq2 vignette.

Does anyone have solution how this time series experiment could be done so that age could be accounted for? Was wondering that would replacing SUBJECT_ID with AGE correct the problem as subject_id is an arbitrary number designated for each individual while age is specific. The study is less than a year so basically the age is constant at each three time point.

Thank you so much in advance!

########################   DIFFERENTIAL ANALYSIS with DESeq2 Case-only #################
########################################################################################
#Lets check that the order is right
coldata <- subset(coldata, coldata$CASE_CONTROL==1)

coldata <- coldata[order(coldata$CASE_CONTROL),]
sample_list <- as.vector(coldata$SAMPLE_ID)
countdata <- countdata[,sample_list]
str(coldata)

#Lets create the modelmatrixes for full and reduced model
#Full
x <- model.matrix(~ PRE_MID_POST + SUBJECT_ID, coldata)
all.zero <- apply(x, 2, function(x) all(x==0))
all.zero

idx <- which(all.zero)
full <- x[,-idx]
full1 <- unname(full)
#Reduced
x <- model.matrix(~ SUBJECT_ID, coldata)
all.zero <- apply(x, 2, function(x) all(x==0))
all.zero

idx <- which(all.zero)
reduced <- x[,-idx]
reduced1 <- unname(reduced)

##################################################################################
############# Lets do the DE with LRT and both modelmatrices #####################
##################################################################################
source("https://bioconductor.org/biocLite.R")
biocLite("DESeq2")
library(DESeq2)

ddsFullCountTable <- DESeqDataSetFromMatrix(countData = countdata,colData = coldata, design = ~1)
ddsFullCountTable <- ddsFullCountTable[ rowSums(counts(ddsFullCountTable)) > 1, ]
keep <- rowSums(counts(ddsFullCountTable) >= 5) >= 5
ddsFullCountTable <- ddsFullCountTable[keep,]
dds <- DESeq(ddsFullCountTable, test="LRT", full=full, reduced=reduced)
DESeq2 DES • 847 views
ADD COMMENT
0
Entering edit mode
@mikelove
Last seen 23 hours ago
United States

Linking to my answer on the other post just now:

Comment: DESeq2 model - How to test case/control time course model when taking account wi

ADD COMMENT

Login before adding your answer.

Traffic: 554 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