length = 3' in coercion to 'logical(1)

[elizabethmcmillan]

Hi! I'm opening another thread about this issue. Can you please let me know if you have any ideas for how to solve this issue?

I'm getting this error when I run both my data and the test data:

library(Rmagic)
data(magic_testdata)
data_MAGIC <- magic(magic_testdata, genes=c("VIM", "CDH1", "ZEB1"))
Error in is.null(x = genes) || is.na(x = genes) :
'length = 3' in coercion to 'logical(1)'

Thank you for your help!!

[leonardfei]

Hi! I'm opening another thread about this issue. Can you please let me know if you have any ideas for how to solve this issue?

I'm getting this error when I run both my data and the test data:

library(Rmagic)
data(magic_testdata)
data_MAGIC <- magic(magic_testdata, genes=c("VIM", "CDH1", "ZEB1"))
Error in is.null(x = genes) || is.na(x = genes) :
'length = 3' in coercion to 'logical(1)'

Thank you for your help!!

Hello, I met this problem recently, and found the error was referred in the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC10789639/), maybe a R version of incompatible problems, but still an error after I modified version. The description in article is as follows:
Problem 1
Error in is.null (x = genes) || is.na(x = genes) : ‘length = 3’ in coercion to ‘logical(1)’ in Step I when finding the optimal cutoff of highly variable EMT genes.

Potential solution
•This issue is related to the incompatibility of the Rmagic package which is one of the dependencies of RCOMET with the new R version 4.3.1+, please use an alternative R version to fix this issue.

Have you solved the problem now?

[AnniceNajafi]

Hi @leonardfei I was an author on the article you have referenced, we created an R package that depends on Magic. I received an email from a user that had encountered a similar issue, so I decided to fork the Magic repo and make changes to fix the bug within the Magic package. As you have noted and we mentioned in our article, the issue is due to the incompatibility of the R package with the new R version. You can read more about the problem here: https://stackoverflow.com/questions/72848442/r-warning-lengthx-2-1-in-coercion-to-logical1

To fix the issue, you can try installing the R package by following the instructions from this forked repo: https://github.com/AnniceNajafi/MAGIC/blob/master/Rmagic/README.md

Here is how you can do that:

Installation from this fork

Step 1: Install the R package

if (!require(devtools)) install.packages("devtools")

#Install Rmagic from this fork
devtools::install_github("AnniceNajafi/MAGIC/Rmagic")

Step 2: Install the Python dependencies

After installing the R package, install the Python magic-impute package:

library(Rmagic)
install.magic()

Important: If you encounter a numpy compatibility error like:

ValueError: numpy.dtype size changed, may indicate binary incompatibility

You need to downgrade numpy to version 1.x. First, find your Python path:

reticulate::py_config()

Then in your system terminal (Command Prompt/PowerShell on Windows, Terminal on Mac/Linux), run:

"<your-python-path>" -m pip install "numpy<2.0" --force-reinstall

Replace <your-python-path> with the python: path shown by py_config().

Step 3: Restart R and test

library(Rmagic)
data(magic_testdata)
data_MAGIC <- magic(magic_testdata, genes = c("VIM", "CDH1", "ZEB1"))
summary(data_MAGIC)

I hope this helps.

Kind regards,
Annice