icbi-lab / infercnvpy

Infer copy number variation (CNV) from scRNA-seq data. Plays nicely with Scanpy.

Home Page:https://infercnvpy.readthedocs.io/en/latest/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Make nextflow pipeline to run copyKAT and SCEVAN

grst opened this issue · comments

I feel rpy2 is unreliable, the R packages are cumbersome to install and a lot of additional dependencies. On top of that, the R methods are relatively slow compared to the infercnv algorithm in Python and it makes sense to run them in parallel on individual patients, potentially on HPC.

I would suggest to

  • remove the copykat function as it is currently from the package
  • build a small nextflow pipeline that
    • takes an anndata object as input
    • splits it up by patient (or whatever variable in obs)
    • converts it to SingleCellExperiment
    • runs copykat and or SCEVAN
    • has the dependencies packaged as docker/singularity containers.
  • provide loader functions for the results of copykat and SCEVAN such that
    the visualizations functions of infercnvpy can be used.

Hello @grst
Any plan to add the tutorial of CopyKat to the website (https://icbi-lab.github.io/infercnvpy/tutorials/reproduce_infercnv.html)?
Thanks!
Best,
YJ