Question

is there a function in edger the same as vst or rlog in deseq2

0

Entering edit mode

linouhao ▴ 20

@linouhao-15901

Last seen 20 months ago

United States

the accessory contains the R script , the test data, the word report, you just need to revise your setwd in the R script.

the script used the pseudo.counts for external analysis, and I also see pseudo.counts numeric matrix of normalized pseudo-counts in your edger user guide.

the word report in the accessory describe it more clearly

Which kind of data in edger can be used for external analysis, such as cox model, heatmap, gene scatter plot difference between groups. Urgent for help.

2.

is there a detaied comparison of your edger with wilcox

edger normalization deseq2 • 2.4k views

ADD COMMENT • link updated 4.5 years ago by Gordon Smyth 52k • written 4.5 years ago by linouhao ▴ 20

1

Entering edit mode

Re: "Urgent":

I discourage users from putting requests for urgency on their posts. Remember you are asking volunteers (who typically are also volunteering in maintaining open source software) for assistance. It is inappropriate to make demands on when it occurs.

ADD REPLY • link 4.5 years ago Michael Love 43k

0

Entering edit mode

sorry, I would not next time

ADD REPLY • link 4.5 years ago linouhao ▴ 20

score 2 · Answer 1 · 2020-06-16

2

Entering edit mode

Kevin Blighe ★ 4.0k

@kevin

Last seen 6 weeks ago

Republic of Ireland

Did you literally just copy-paste this from an exam? If you are not interested in what is exactly happening in the underlying calculations, then the cpm() function would be the general equivalent in EdgeR. Pay close attention to the log and prior.count parameters.

Kevin

ADD COMMENT • link 4.5 years ago Kevin Blighe ★ 4.0k

0

Entering edit mode

sorry， I ever tried to send the author an email, the content is the same. thanks a lot.

log logical, if TRUE then log2 values are returned. prior.count average count to be added to each observation to avoid taking log of zero. Used only if log=TRUE.

so what is the recommened parameter suggested for cpm, use the default?

the following is four kind of counts data, can you distinguish it, I am coonfused, especially the pseudo.counts? is this also the same function as vst in deseq2

https://sbimg.cn/image/0t7Wd

https://sbimg.cn/image/0tHV4

https://sbimg.cn/image/0tW6Y

https://sbimg.cn/image/0tZbV

ADD REPLY • link 4.5 years ago linouhao ▴ 20

score 2 · Answer 2 · 2020-06-17

2

Entering edit mode

Gordon Smyth 52k

@gordon-smyth

Last seen 8 hours ago

WEHI, Melbourne, Australia

Dear linouhao,

I'll repeat here what I told you by email:

The cpm() function edgeR converts counts to numerical values that can be analysed using external statistical programs, see Section 2.16 in the edgeR User's Guide.
edgeR is always more powerful and correct than the Wilcoxon test, for any number of samples.

The documentation says not to use pseudo counts for external analyses. Everyone (documentation, me, Kevin) has told you the same thing.

ADD COMMENT • link 4.5 years ago Gordon Smyth 52k

0

Entering edit mode

sorry for disturbing thanks a lot. so what pseudo counts can be used for

ADD REPLY • link 4.5 years ago linouhao ▴ 20

1

Entering edit mode

See Section 2.8.7 of the edgeR User's Guide.

ADD REPLY • link 4.5 years ago Gordon Smyth 52k

0

Entering edit mode

got it. thanks. because someone in some course used this. so after cpm, the output ca be used for heatmap, cox and so on? am I right?

ADD REPLY • link 4.5 years ago linouhao ▴ 20

1

Entering edit mode

Correct, generally. Try to become familiar with data distributions and regularly plot your data via hist(). Most downstream functions expect data to follow a normal distribution, or that the residuals follow a normal distribution, whichever is appropriate. You will see a vast difference between the distributions of raw/normalised RNA-seq counts and transformed RNA-seq counts.