Is there other IDE like R studio working on the linux?
5
1
Entering edit mode
10 weeks ago

Hi, every one

I want to run multi version R on linux, because I want to test different bioconductor package version. but R studio only support one version at a time. And I can not afford R studio pro. So I am wondering whether some one can recommend some other IDE for R in bioinformatics anlysis :).

Best wishes

Guandong Shang

Rstudio R bioinformatics bioconductor • 298 views
ADD COMMENT
1
Entering edit mode

There's always Emacs..

ADD REPLY
0
Entering edit mode

hhh, I have decided to use vscode-R :)

ADD REPLY
5
Entering edit mode
10 weeks ago
ATpoint 47k

The simplest solution would probably be to install multiple versions of R(Studio) via a package manager such as conda into separate environments created with conda create. You should probably ask yourself though whether such an endeavor is really what your analysis needs (I assume you have trouble somewhere), and whether it is worth the bother rather than figuring out what the true underlying cause of the troubles are beyond package versions.

ADD COMMENT
0
Entering edit mode

Thanks, Dr Toenges :). Actually, the balance between update and matain alwasy confuses me. For example, I have know some bugs in the package of bioconductor and the pakcages have fixed the bug in the newest version. I am not sure whether I should update R for the newest package or I just use the old version.

ADD REPLY
0
Entering edit mode

Ah, I see. Would you need to upgrade your R and Bioconductor version as well? This then of course would mean that other packages get upgrades as well, making it difficult to exactly reproduce results from previous analysis. I guess a separate environment would in fact be the simplest. There are other ways to make separate environments as well, e.g. venv/virtualenv (python).

ADD REPLY
0
Entering edit mode

Yes, my main concern is bioconductor version :), which is depends on the R version╮(╯_╰)╭

ADD REPLY
0
Entering edit mode

You could consider using something like packrat (I believe there's a newer "model" called renv) - this would isolate r and versioned packages per project, which would work better with R Studio as RStudio should pick up the appropriate R binary when you open a project.

EDIT: On reading the renv docs, it's not clear if it supports multiple R versions. I'd still give it a shot though.

ADD REPLY
3
Entering edit mode
10 weeks ago

I believe you can install multiple R kernels in Jupyter. Another solution is to use RStudio out of containers, one for each version you need.

ADD COMMENT
0
Entering edit mode

Thanks, I will try jupyter :)

ADD REPLY
2
Entering edit mode
9 weeks ago
bernatgel ★ 2.9k

The Linux system in my institution have several R versions and several Rstudio versions running in parallel. This is managed by the SysAdmin, but I think it's "simply" a question of installing them on different directories. In addition, I'm pretty sure you can set an environment variable before starting Rstudio to tell it what version of R to use, so you can have multiple instances running on different R versions at the same time.

ADD COMMENT
0
Entering edit mode
9 weeks ago
fracarb8 ▴ 130

The best way would be to install different versions of the same package in different libraries and switch between them as you need them

# read packages from library 1

library(myPackage, lib.loc="~/Rlib/v1/myPackage")

# read packages from library 1

library(myPackage, lib.loc="~/Rlib/v2/myPackage")

ADD COMMENT
0
Entering edit mode

How would you switch between R versions using your approach?

ADD REPLY
0
Entering edit mode

In this specific case this is (I guess) not possible as OP is using Bioconductor. Bioc package versions are tied to Bioc versions itself which again are tied to R versions. One would need to explicitely use a different R version prior to your suggestion, any idea for that?

ADD REPLY
0
Entering edit mode

BiocManager allows you to install "local" packages. You could download all the different version from bioconuctor and install them using different location (e.g setting different Rlibs) A quite convoluted way, but possible....I think.

ADD REPLY
0
Entering edit mode
5 weeks ago

I think https://github.com/Ikuyadeu/vscode-R may be most suitable for me :)

ADD COMMENT

Login before adding your answer.

Traffic: 1778 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