Question

How to merge two tximport object into one dataset for DESeqDataSetFromTximport

0

Entering edit mode

rensanhao • 0

@rensanhao-21464

Last seen 5.5 years ago

Hi,

I created two the tximport object (untreated vs treated) from salmon,which is a S4 object has four cols name as abundance, counts, length, and countsfromabundance.My question is how to merge these two dataset into one so I could import them with using DESeqDataSetFromTximport?

Thanks.

deseq2 tximport • 2.4k views

ADD COMMENT • link updated 5.5 years ago by Michael Love 43k • written 5.5 years ago by rensanhao • 0

score 1 · Answer 1 · 2019-07-26

1

Entering edit mode

Michael Love 43k

@mikelove

Last seen 2 hours ago

United States

I'd recommend this approach:

files <- c(files1, files2)
txi <- tximport(files)

Otherwise, if you had used tximeta you could simply cbind the SummarizedExperiment objects.

For the list output of tximport, you just need to cbind each of the matrix elements.

ADD COMMENT • link 5.5 years ago Michael Love 43k

0

Entering edit mode

Thanks for this quick reply. So following this merged dataset, how to label which group is my control group or treated group? or should I add a column to label the salmon-quant.sf files before this integration?

ADD REPLY • link 5.5 years ago rensanhao • 0

0

Entering edit mode

Add it in the dds construction step (eg DESeqDataSetFromTximport) as colData.

ADD REPLY • link 5.5 years ago Michael Love 43k

0

Entering edit mode

I'm sorry but I didn't get it. According to vignette,I think the code

files <- file.path(dir,"salmon", samples$run, "quant.sf.gz")
names(files) <- samples$run

is essential for labeling the group info, but I didn't fully understand how it works and moreover, it also led me to confuse that

> dsTxi <- DESeqDataSetFromTximport(txi,
>                                    colData = samples,
>                                    design = ~ condition)

the design arguments should also work for specifying the group information? I just started learning R and RNA-seq so I'll very appreciate if you feel necessary to explain some basic knowledge regarding this issue.

Thanks a lot.

ADD REPLY • link 5.5 years ago rensanhao • 0

0

Entering edit mode

Maybe you can partner with a person who has experience with R. On the support site I try to assist users who have questions about the software, but you probably need to have some R background to get started. You want to have a table of sample information so that you can import all the samples at once with one call to tximport().

ADD REPLY • link 5.5 years ago Michael Love 43k

0

Entering edit mode

Thanks. I have some clues about this now.

ADD REPLY • link 5.5 years ago rensanhao • 0

0

Entering edit mode

Thanks for this quick reply. So following this merged dataset, how to label which group is my control group or treated group? or should I add a column to label the salmon-quant.sf files before this integration?

ADD REPLY • link 5.5 years ago rensanhao • 0