RangedSummarizedExperiment in Deseq2 and DESeqDataSetFromMatrix
1
0
Entering edit mode
@lirongrossmann-13938
Last seen 22 months ago

Hi All,

In the Deseq2 vignette the first step is creating the object

dds <- DESeqDataSet(se, design = ~ batch + condition)

based on the RangedSummarizedExperiment object se.

In my case, I have a raw count expression matrix (that I got from Kallisto) stored in "exp" and the metadata stored in "met".

I used the following code without creating  RangedSummarizedExperiment object:

mydds <-DESeqDataSetFromMatrix(countData = ecp,colData = met,design = ~batch+condition)

Is mydds == dds?

Thanks!

0
Entering edit mode

is your count data stored in exp or ecp?

0
Entering edit mode

exp.... sorry for the typo

0
Entering edit mode
Assa Yeroslaviz ★ 1.5k
@assa-yeroslaviz-1597
Last seen 17 days ago
Germany

that is correct. try str(dds ) from the example and than also str(mydds) from your data and see if they have the same strucutre.

but to make your life easier Michael Love already created a package to import Kallisto data into DESeq2. A: Count matrix from Kallisto as input for DESeq2

0
Entering edit mode

Thank you!

Does it matter if I use the count matrix I have  (the one I got from Kallisto) with Deseqdatafrommatrix instead of tximport? My understanding is that I need to use tximport if I have transcript abundance estimates, but I have reads counts per genes, i.e. counts of reads aligned to a gene.

Would it make any difference in terms of the DEG I got?

Thanks