How to index design formula elements in DESeq2?

Question

Here is an example:

cts<-matrix(sample(1:1000, 600, replace = T), ncol=6)
rownames(cts)<-paste0("Gene", 1:100)
colnames(cts)<-paste0("sample", 1:6)

coldat<-data.frame(B=c(1,2,1,2,1,2),
                   C=c(1,1,1,2,2,2))
rownames(coldat)<-colnames(cts)

library(DESeq2)
batch="B"
condition="C"
dds <- DESeqDataSetFromMatrix(countData = cts,
                              colData = coldat,
                              design= ~ batch + condition)
dds <- DESeq(dds)
Error in DESeqDataSet(se, design = design, ignoreRank) : 
all variables in design formula must be columns in colData

My question is how to index batch="B" and condition="C" in the design formula.

score 2 · Accepted Answer · answered Jan 18 '20 at 15:31

2

Just figured out:

dds <- DESeqDataSetFromMatrix(countData = cts,
                              colData = coldat,
                              design= as.formula(paste0("~", batch, "+", condition)))

answered Jan 18 '20 at 15:31

David Z

6,641
11
50
101

If this has worked, accept as answer. – zx8754 Jan 21 '20 at 15:21

fra · Answer 2 · 2020-01-18T15:39:15.630

1

The issue is that batch and condition are not esplicitly declared in coldat. If B=batch and C=condition, then you can simply ~ B+C. Otherwise you will need to add (or rename) the 2 columns.

coldat$batch <- coldat$B
coldat$condition <- coldat$C

or, if they differ

coldat$batch <- c(1,2,3,4,5,6)
coldat$condition <- c(1,1,1,2,2,2)

edited Jan 18 '20 at 15:39

answered Jan 18 '20 at 15:23

fra

832
6
14

How to index design formula elements in DESeq2?

2 Answers2