search

hillarykoch / powerTCR

R package. Main goals are to fit models to the clone size distribution of the TCR repertoire, and to perform model-based comparative analysis of samples.
Other
5 stars 2 forks source link

tcR package will soon be orphaned #2

Closed vadimnazarov closed 4 years ago

vadimnazarov commented 4 years ago

Hi @hillarykoch ,

thank you for supporting the RepSeq community with your package! I would like to notify you that the tcR package will soon be orphaned in favour of our new package immunarch - https://github.com/immunomind/immunarch

I would be glad to help you move to the immunarch dependency.

What parts of tcR do you use?

hillarykoch commented 4 years ago

Hi @vadimnazarov ,

Thanks for letting me know about this. My usage of tcR is fairly minimal, and contained entirely in the powerTCR/R/file_readers.R file. I use parse.file(), parse.folder(), and get.inframes()

I would appreciate any help! And will check out immunarch.

vadimnazarov commented 4 years ago

@hillarykoch , thank you for the prompt response!

Functions parse.file() and parse.folder() are combined into the singular function repLoad that automatically detects the input file format and always returns a list with two sub-lists data and meta. data is a list of repertoires. The data format for immunarch is a little bit different, feel free to take a look here: https://immunarch.com/articles/v2_data.html

get.inframes is now just coding.

I hope this will help you to get started, and please don't hesitate to ping me here. I will monitor this issue until it's closed.

Again, thank you a lot for your contribution!

hillarykoch commented 4 years ago

Right, hopefully just one quick question from me-- I am wondering what the structure is of the object returned from repLoad? I am only trying to extract the counts from the output of this file reader. So, currently, I have something like

data <- tcR::parse.file(.filename = file, .format = format)
counts <- data$Read.count

I don't have any of the rep-seq data sitting around to play with to test this. Would the above code (assuming I change it to immunarch::repLoad) be correct in this case?

Thanks so much.

vadimnazarov commented 4 years ago

No, but it's easy to overcome. The immunarch code for this exact logic would be:

data <- immunarch::repLoad(.path = file)$data[[1]]
counts <- data$Clones

Note that there is no need to provide the .format argument.

vadimnazarov commented 4 years ago

Hi @hillarykoch , the immunarch is on CRAN now so feel free to test if powerTCR works!