Question: Get release date for version of DESeq2
0
gravatar for thibault.lorin34
12 months ago by
thibault.lorin3420 wrote:

Hi,

For some reason I cannot reproduce the results I obtained using DESeq2 around one year ago (the main conclusions are the same, with the same top hits, but the results are marginally changed when it comes to DEGs at the "bottom" of the list). I thought (among many other options) that it might be because of a new DESeq2 installation since then that I would have done without remembering it.

The package I'm currently using is:

> packageVersion("DESeq2")

[1] ‘1.16.1’

I obtained the release date of such package here. In the vignette it is indicated that the date is October 17, 2017. Does this mean that this version corresponds to that date? Can there be multiple vignettes for a single version?

Thanks for your help,

Thibault

deseq2 R • 394 views
ADD COMMENTlink modified 12 months ago by Michael Love24k • written 12 months ago by thibault.lorin3420
Answer: Get release date for version of DESeq2
2
gravatar for Michael Love
12 months ago by
Michael Love24k
United States
Michael Love24k wrote:

You want the release date for the version that you used previously? The vignette date may or may not help because if you are using an out of date version of R you would not have obtained eg the 2017 version of Bioconductor and DESeq2. There’s not a great way to figure this out unless you saved the dataset object or put sessionInfo() at the end of your reports (I follow this practice and recommend it in the workflow).

ADD COMMENTlink written 12 months ago by Michael Love24k

I would like to get the release date of the version I'm currently using (1.16.1): does it correspond to the "vignette date" (Oct. 17th 2017)? The R version I am using is R version 3.4.2 (2017-09-28)

ADD REPLYlink modified 12 months ago • written 12 months ago by thibault.lorin3420
1

Oh, I see. Yes if you find the version of Bioconductor online that gives you DESeq2 1.16, then it will say the date.

Also the pattern is: current version of DESeq2 is v1.20, released 4/2018. Previous even versions are released 6 months apart. So v1.18 is from 10/2017, and v1.16 is from 4/2017, etc. back to v1.0 released in 4/2013.

ADD REPLYlink written 12 months ago by Michael Love24k
1

Also, the 'landing pages' are accessible by Bioconductor version number, so https://bioconductor.org/packages/3.4/bioc/html/DESeq2.html takes one to the DESeq2 landing page for the 3.4 release of Bioconductor (note the '3.4' in the url); generally one could get there by going to 'About' then 'Release Announcements' then clicking on the 3.4 release for 'biocViews' corresponding to the 3.4 release. Also starting from Bioconductor 3.6 all release versions of packages were archived at https://bioconductor.org/packages/3.6/bioc/src/contrib/Archive/ (the 'final' release version is accessible from the package landing page).

ADD REPLYlink written 12 months ago by Martin Morgan ♦♦ 23k
Thanks! Just to keep you updated with this: I switched R version from `3.4.2` to `3.3.2` using the RSwitch utility, and ran my script again. I obtain the same result as was obtained last year \o/ (quite reassuring actually). So the DESeq2 version was indeed the cause of the slight discrepancy between my "old" and "new" results. Thanks again!
ADD REPLYlink written 12 months ago by thibault.lorin3420

Could you tell me if the following is correct?

If I understand correctly, each version of Bioconductor corresponds to an R version. For instance, Bioconductor versions 3.5 and 3.6 correspond to R version 3.4.

Hence, this means that if my R version is 3.4.2 and was installed on Sept. 28th 2017, I can only have DESeq2 versions that correspond (or are posterior) to the Bioconductor 3.5 release (so, posterior to 1.16.1). That also means that if I my initial results were obtained in August 2017 using R version 3.3.2 it can only have been using at most DESeq2 1.14.1... so that the DESeq2 version I'm using today is different from the one I used last year.

ADD REPLYlink modified 12 months ago • written 12 months ago by thibault.lorin3420
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 283 users visited in the last hour