Week #1 - General Discussion Thread

[DavidRach]

With all three live-streams this week now complete, opening this general discussion thread for "Cytometry in R - Week 1 - Installing R packages" for anyone wanting to discuss any of the topics, or seeking further clarification.

[MLGRANT]

I can't figure out how to find where the homeworks are - I can see on Positron the image of where they should be but the left sidebar doesn't match mine - I'm missing the DavidRach folder. Please can you guide me how to get the folder?

[DavidRach]

Hi @MLGRANT, Homeworks are the optional take-home problems portion of each weeks content, located right after the Additional Resources section.

The home-works folder would be in the course folder, within the respective week. This would be where you would create your own folder and attach anything being turned in (see here for details)

Best-
David

[Algarues]

the instruction about getting help on sessioninfo is inaccurate

If I type wherever sessioninfo()<

Positron will not recognize it as a function. Most likely, it needs to be put between the entries that indicate the beginning and end of code.

However, if I hoover on any of the already present sessioninfo() functions, the help will work.

Thus, the walkthrough should include instructions on how to correctly state sessioninfo() in the file.

[DavidRach]

I think what you are seeing is working within a .qmd file, Positron treats text like text, and code like code. Consequently, you only get highlighting within code blocks when working with a .qmd file.

However, if you are working in an .R file, it immediately retrieves, because everything within those files is treated like code.

[Algarues]

Thank you for replying.
Since the walkthrough is based on the .qmd file, it does not make much sense to explain a program behaviour in another type of file that is not being used.

Mentioning the difference is critical in this context. Otherwise the instructions are simply unclear.

Hope it makes sense.

[DavidRach]

Noted, the above was mainly providing you some additional context about the odd behavior. We will be using both file types extensively throughout the course going forward.

[Algarues]

When installing packages on Positron for Windows there is the warning that the library is not writable and the option to install on personal library is offered instead.

I guess the suggested choice is to use the personal library. Correct?

[DavidRach]

From my recollection of the time I tested it, it would provide the permissions for write, but would likely require opening Positron as administrator continuously going forward with each additional install. Personal library is likely the less-hassle route long-term if the computer is primarily used by one person.

[Algarues]

Thank you

I was wondering that installing the whole thing inside Windows subsystem for linux may be feasible too....
;)

[DavidRach]

I was wondering that installing the whole thing inside Windows subsystem for linux may be feasible too.... ;)

Screenshot and blog the process if you do 🤣

[Algarues]

Regarding the "Your turn" section of lesson 1
I saw that many packages were loaded and not attached like purrr flowCore etc.

Thus, I made them available straightaway with library()

[Algarues]

is there some type of "grep" that can be piped after sessioninfo?

[kash-jm]

Hi I am having trouble with the last bit of the homework. I'm trying to install the last package from github using pak, but keep getting this error. not sure how to go about this

[DavidRach]

Hi @kash-jm, the last question was the intentionally tricky one, as both superheat and EmbedSOM packages will fail when trying to install via pak. If you want a hint, click on the details arrow below. Best- David

Details

First figure out for each problem dependency package, figure out which repository they are currently in.
Then use correct install command to install each of them.
Once all dependency conflicts are addressed, try to pak install CytoExploreR once again.

[kash-jm]

Thanks for the hint @DavidRach managed to figure it out!

[Oscar-Kai]

I am still in week1 stuck at why I can not render the .qmd to pdf. The html works below is a screen short of the output from Positron Terminal on the qmd file.

PS C:\Users\okai\Documents\R Project Folder\GitPractice2> quarto render QuartoTest.qmd to -- pdf

processing file: QuartoTest.qmd
1/3
2/3 [unnamed-chunk-1]
3/3
output file: QuartoTest.knit.md

pandoc to
to: html
output-file: QuartoTest.html
standalone: true
section-divs: true
html-math-method: mathjax
wrap: none
default-image-extension: png
variables: {}

metadata
document-css: false
link-citations: true
date-format: long
lang: en

pandoc: to: withBinaryFile: does not exist (No such file or directory)
WARN: Error encountered when rendering files
PS C:\Users\okai\Documents\R Project Folder\GitPractice2>

[UMGCCCFCSR]

Hello @Oscar-Kai, the follow-up question from last week were on the GitHub personal access token thread, it was in reply to the answer you referenced.

We provided the list of software needed to be installed and during the first week material the list of R packages. The issues you are encountering is because Git and Quarto may not have entirely installed properly for whatever reason. This might be due to IT admin policy on the local computer, or it may not. Regardless, try the steps suggested in last week's reply, and I will attempt to help troubleshoot. If it is easier for your IT to do it, then take that approach.

Best Wishes-
David

[Oscar-Kai]

@DavidRach the ones indicated those i installed. did not have any adm issue with our institutional policy. But some how for whatever reasons i could not render the qmd file to the extension mentioned.

For instance to day the toc argument is not working. I was trying to do the table of content but i must be missing something something the header looks like typed title header

[Oscar-Kai]

@DavidRach I got it. I was not putting the dashes instead they lood like bacticks to me. I am good

[Oscar-Kai]

@DavidRach now something else has come up i cant silence the errors even when #| warnings: FALSE & #| messages: FALSE

[DavidRach]

Hi @Oscar-Kai , glad the first issue was fixed. In R, you go from messages to warnings to errors. In the case of the above, there is no installed ggcyto package, so an error is thrown.

If you want to have a chunk of code that contains a known error (for a demonstration for example) set

#| error: TRUE

Alternatively, you can install the R package, and the error should go away

library(BiocManager)
install("ggcyto)
# See Installing R packages Week 1 for details

Best-
David