class name has no 'package' attribute

[supermaxiste]

Hello,

I've been stuck for a while on an error I'm getting when running the file 20181125-filtered_simulated_data_dmrseq.Rmd. The following line is causing the error:

bs <- BSseq(
  chr = tab$chr, pos = tab$pos,
  M = tab[, grep(pattern = ".M", x = colnames(tab))],
  Cov = tab[, grep(pattern = ".cov", x = colnames(tab))],
  pData = anno, sampleNames = anno$names
)

When I run it I get the following:

Error in validMsg(msg, .Call(cxx_check_M_and_Cov, M, Cov)) : 
  class name has no 'package' attribute

I tried to look for solutions but the few things I tried didn't seem to work. Could you please check if you get the same error?

Here's a list of all the tools I installed on conda for running this script:

  - r-base=3.6.2
  - bioconductor-geoquery=2.54.0
  - r-rmarkdown=2.1
  - java-jdk=8.0.112
  - r-xtermstyle=3.0.5
  - pigz=2.3.4
  - r-devtools=2.0.2
  - bioconductor-dmrseq=1.6.0
  - r-data.table=1.12.8
  - r-tidyverse=1.3.0
  - r-dplyr=0.8.3

[dktanwar]

Hi @supermaxiste

The error I got is:

Error in validMsg(msg, .Call(cxx_check_M_and_Cov, M, Cov)) : 
  data.frames should be converted to matrices

Because of this error, I added data.matrix to line 64 and 65 in: Rmd file

I think that this has something to do with package versions. Here are the package versions I am using:

version  R version 3.6.2 (2019-12-12)

 bsseq                  * 1.22.0     2019-10-29 [1] Bioconductor                       
 dmrseq                 * 1.7.3      2020-01-14 [1] Github (kdkorthauer/dmrseq@437f94f)

[dktanwar]

Sorry, pushed to master branch by mistake!

[dktanwar]

Updating bsseq should solve the issue. Let me know!

[supermaxiste]

Thank you for the quick response @dktanwar, I'll check it out tomorrow!

[supermaxiste]

Merged to master by mistake, will revert changes, sorry

[supermaxiste]

Issue fixed!