Question

how to handle missing data for SampleTable in DESeq2

0

Entering edit mode

anpham ▴ 60

@anpham-7402

Last seen 7.2 years ago

I have RNAseq data for human subjects and plan to do a differential gene expression analysis using DESeq2. The covariates I plan to use include sex, age, genotypic principal components, and substance use. However, I have missing data for principal components and substance use and would like to know how to code these missing data in the SampleTable. Can I put "." for missing data or would I have to take out the subjects with missing data? Thanks.

deseq2 sample_table missing data • 3.1k views

ADD COMMENT • link updated 5.8 years ago by g.wang2 • 0 • written 9.7 years ago by anpham ▴ 60

score 0 · Answer 1 · 2015-03-29

0

Entering edit mode

Michael Love 42k

@mikelove

Last seen 6 hours ago

United States

Hi, Unfortunately we don't have any support for inference on covariates with missing data. You'd have to remove those subjects.

ADD COMMENT • link 9.7 years ago Michael Love 42k

score 0 · Answer 2 · 2019-01-23

0

Entering edit mode

g.wang2 • 0

@gwang2-19579

Last seen 2.4 years ago

United Kingdom

Hello there,

I found this post tends to overlap with the question I'd like to ask, but not entirely. Therefore, adding the question here to this post:

There're some low quality RNA-seq data or some data looking dissimilar in the sample distance analysis from a couple of samples that need removal. The samples were collected before, during and after a treatment. The question is - should the problematic samples be excluded only or the entire subject data to be removed in DESeq2 analysis? Thanks.

Regards Guan

ADD COMMENT • link 5.8 years ago g.wang2 • 0

0

Entering edit mode

If you have a question, start a new thread. Simply piggy-backing on some four year old thread that you think is similar enough isn't the way to go. In addition, don't use the Add Answer button to ask questions! That button is literally wrapped in instructions saying that you should only use it to answer a question, so why would you use it to ask another question?

ADD REPLY • link 5.8 years ago James W. MacDonald 67k

0

Entering edit mode

I see, thanks for pointing these out. I've now started a new post asking the question.

Regards Guan

ADD REPLY • link 5.8 years ago g.wang2 • 0