# Problem with R kernel: ‘Matrix (\>= 1.5.0)’ not available

**URL:** <https://renku.discourse.group/t/problem-with-r-kernel-matrix-1-5-0-not-available/838>\
**Category:** General\
**Created:** [30 June 2023 09:44 UTC](https://renku.discourse.group/t/problem-with-r-kernel-matrix-1-5-0-not-available/838 "2023-06-30T09:44:41Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![alexdray86](https://yyz2.discourse-cdn.com/free1/user_avatar/renku.discourse.group/alexdray86/32/631_2.png) [@alexdray86](https://renku.discourse.group/u/alexdray86)\
**Post date:** [30 June 2023 09:44 UTC](https://renku.discourse.group/t/problem-with-r-kernel-matrix-1-5-0-not-available/838/1 "2023-06-30T09:44:41Z")

</div>

Dear all,

I am trying to install a list of R packages to be used on a notebook with an R kernel.

I cannot install this package from gtihub:

devtools::install\_github( “Danko-Lab/BayesPrism/BayesPrism”)

Among the warning and errors, it seems that the first popping up is ‘Matrix (\>= 1.5.0)’ not available, which then seems to generate error in multiple libraries.

My reading on the topic regarding this error is that R has to be updated, but I might be wrong. The package has a lot of dependencies and the error gets triggered at some point.

Would you have any advice on how to approach the problem ? Here is the full output that I am getting when I try to install the packages:

Downloading GitHub repo Danko-Lab/BayesPrism@HEAD

formatR (NA → 1.12 ) [CRAN]  
futile.op… (NA → 1.0.1 ) [CRAN]  
lambda.r (NA → 1.2.4 ) [CRAN]  
snow (NA → 0.4-4 ) [CRAN]  
futile.lo… (NA → 1.4.3 ) [CRAN]  
RcppHNSW (NA → 0.3.0 ) [CRAN]  
BiocNeigh… (NA → 1.16.0 ) [CRAN]  
BiocParallel (NA → 1.32.6 ) [CRAN]  
beachmat (NA → 2.14.2 ) [CRAN]  
rsvd (NA → 1.0.5 ) [CRAN]  
ScaledMatrix (NA → 1.6.0 ) [CRAN]  
sparseMat… (NA → 1.10.0 ) [CRAN]  
locfit (NA → 1.5-9.5 ) [CRAN]  
DelayedMa… (NA → 1.20.0 ) [CRAN]  
registry (NA → 0.5-1 ) [CRAN]  
metapod (NA → 1.6.0 ) [CRAN]  
bluster (NA → 1.8.0 ) [CRAN]  
BiocSingular (NA → 1.14.0 ) [CRAN]  
statmod (NA → 1.4.36 ) [CRAN]  
edgeR (NA → 3.40.2 ) [CRAN]  
scuttle (NA → 1.8.4 ) [CRAN]  
gridBase (NA → 0.4-7 ) [CRAN]  
rngtools (NA → 1.5.2 ) [CRAN]  
pkgmaker (NA → 0.32.2 ) [CRAN]  
scran (NA → 1.26.2 ) [CRAN]  
NMF (NA → 0.24.0 ) [CRAN]  
snowfall (NA → 1.84-6.1) [CRAN]  
Installing 27 packages: formatR, futile.options, lambda.r, snow, futile.logger, RcppHNSW, BiocNeighbors, BiocParallel, beachmat, rsvd, ScaledMatrix, sparseMatrixStats, locfit, DelayedMatrixStats, registry, metapod, bluster, BiocSingular, statmod, edgeR, scuttle, gridBase, rngtools, pkgmaker, scran, NMF, snowfall

Warning message:  
“dependency ‘Matrix (\>= 1.5.0)’ is not available”  
Warning message in i.p(…):  
“installation of package ‘DelayedMatrixStats’ had non-zero exit status”  
Warning message in i.p(…):  
“installation of package ‘scuttle’ had non-zero exit status”  
Warning message in i.p(…):  
“installation of package ‘scran’ had non-zero exit status”

- checking for file ‘/tmp/Rtmp6tQ3CV/remotes16b72b1d6904/Danko-Lab-BayesPrism-1ad3e82/BayesPrism/DESCRIPTION’ … OK
- preparing ‘BayesPrism’:
- checking DESCRIPTION meta-information … OK
- checking for LF line-endings in source and make files and shell scripts
- checking for empty or unneeded directories
- building ‘BayesPrism\_2.0.tar.gz’

Warning message in i.p(…):  
“installation of package ‘/tmp/Rtmp6tQ3CV/file16b7302681db/BayesPrism\_2.0.tar.gz’ had non-zero exit status”

Thank you very much !

Best,

Alexandre Coudray  
PhD student at EPFL

---

<div class="post-metadata">

**Author:** ![elisabetc](https://yyz2.discourse-cdn.com/free1/user_avatar/renku.discourse.group/elisabetc/32/811_2.png) [@elisabetc](https://renku.discourse.group/u/elisabetc)\
**Post date:** [14 July 2023 07:53 UTC](https://renku.discourse.group/t/problem-with-r-kernel-matrix-1-5-0-not-available/838/2 "2023-07-14T07:53:45Z")

</div>

Thank you @alexdray86 for posing your question. The R version in the current Renku templates is about the same time the Project **[BayesPrism](https://github.com/Danko-Lab/BayesPrism)** was created. Renku precisely aims to solve the problem that you are currently facing (improve code reusability by a containerized compute environment). Could you test locally with a newer version of R to check if the problem persists?