Normalizing Salmon output count data matrix using DESeq2
1
0
Entering edit mode
kiqbal • 0
@7cabb0d9
Last seen 15 months ago
United States

I import counts and abundance matrix from Salmon output using tximport. I want to normalize the Salmon output count matrix using DESeq2 package. What code should I use to achieve this task? Can I do similar with Abundance output of Salmon?

Thank you

DESeq2 tximport counts Salmon • 1.3k views
ADD COMMENT
2
Entering edit mode
@mikelove
Last seen 1 hour ago
United States

Do you mean scale the counts for sequencing depth?

dds <- DESeqDataSetFromTximport(txi, ~condition)
ncts <- counts(dds, normalized=TRUE)

The abundance isn't something that DESeq2 operates on (e.g. statistics are made from counts alone).

If you want to scale TPM using a similar procedure of median-ratio, you can manually calculate TPM size factors with estimateSizeFactorsForMatrix.

ADD COMMENT
0
Entering edit mode

Just out of curiosity, what type of normalization is applied by the "counts(dds, normalized=TRUE)" statement? I mean, is there a specific name for this normalisation? I want to know about the mathematics behind this. Thank you.

ADD REPLY
0
Entering edit mode

Median ratio scaled counts. Read the DESeq2 paper for the median ratio calculation.

If you are using tximport, note that we use a different approach, described in tximport paper.

ADD REPLY
0
Entering edit mode

Thank you for the reference. Also, if I am using tximport and then normalising the counts, at the backend the normalising (which is specific to counts from tximport, as you mentioned is different) is automatically taken care of? Right?

ADD REPLY
0
Entering edit mode

Yes, tximport --> DESeq2 corrects for sequencing depth in addition to gene length.

ADD REPLY

Login before adding your answer.

Traffic: 660 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