broadinstitute/celligner

tumor - cancer cell line alignment. Use it on the depmap portal or install it with pip.

celligner's Issues

mnnpy: error building wheel

I have successfully been able to install Celligner in both Python and R, however, when attempting to run it, the package mnnpy gives many errors building the wheel. I have tried using brew, cloning the mnnpy repository, I tried using gcc, llvm, lomp, and libomp, all unsuccessful on two different macs and a PC, all with no success at installing mnnpy. Is anyone able to help with this issue?

__init__() got an unexpected keyword argument 'low_memory'

Hi team,

Thank you for developing this powerful tool. We are trying to use it for our own data but it hit this error when I tried to run the .transform().

reducing dimensionality...
doing differential expression analysis on the clusters..
running differential expression on 27 clusters
running limmapy on the samples
you need to have R installed with the limma library installed
3.5.3
there is 0.283 overlap between the fit and transform dataset in their most variable genes
doing cPCA..

TypeError Traceback (most recent call last)
/var/folders/7r/70gsq7hd1p5f6d8cgzf0ftb80000gr/T/ipykernel_43417/3434172114.py in
----> 1 _ = my_alligner.transform(X_val_one_percentile)

~/anaconda3/envs/local/lib/python3.9/site-packages/celligner/init.py in transform(self, X_pression, annotations, only_transform, _rerun, _doCPCA, recompute_contamination)
580 # TODO: try the automated version, (select the best alpha above 1?)
581 self.cpca_loadings = (
--> 582 CPCA(
583 standardize=False,
584 n_components=self.cpca_ncomp,

TypeError: init() got an unexpected keyword argument 'low_memory'

Looking at the code, it passes this argument when creating the CPCA instance.
https://github.com/broadinstitute/celligner/blob/master/celligner/__init__.py#L585

However, checking the contrastive package, it seems it doesn't accept this argument even for their latest version on Github.
https://github.com/abidlabs/contrastive/blob/master/contrastive/__init__.py#L44

Could you please advise what I should do in this case? Thank you very much in advance.

ValueError: annotations do not match X_pression

Hi,

Thanks for making the tool available.
I am trying to run it but I am struggling a bit. As I do not have access to the test datasets, I do not know how the annotation files are supposed to look like. I think my expression and annotation matrices have the same size, but when I try to run the .fit() function I get this error. Any ideas why? thanks.

Installation gives error

installing with devtools::install_github("broadinstitute/celligner/R") give error:

Error: package or namespace load failed for  celligner  in namespaceExport(ns, exports):
 undefined exports: .average_correction, .center_along_batch_vector, .compute_tricube_average, .tricube_weighted_correction, calc_gene_stats, calc_tumor_CL_cor, check_NAs, cluster_data, create_Seurat_object, find_differentially_expressed_genes, get_cluster_averages, load_additional_data, load_data, modified_mnnCorrect, run_Celligner, run_MNN, run_cPCA, run_cPCA_analysis, run_lm_stats_limma_group, run_multidataset_alignment
Error: loading failed
Execution halted
ERROR: loading failed

sessionInfo()

R version 4.2.1 (2022-06-23)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.5 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/libf77blas.so.3.10.3
LAPACK: /company/software/lapack/3.6.0/lib/liblapack.so

locale:
 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8       
 [4] LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8   
 [7] LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C          
[10] LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C   

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
 [1] ps_1.7.2          prettyunits_1.1.1 rprojroot_2.0.3   crayon_1.5.2     
 [5] withr_2.5.0       R6_2.5.1          lifecycle_1.0.3   magrittr_2.0.3   
 [9] rlang_1.0.6       cachem_1.0.6      cli_3.4.1         curl_4.3.3       
[13] remotes_2.4.2     fs_1.5.2          callr_3.7.3       ellipsis_0.3.2   
[17] devtools_2.4.3    tools_4.2.1       glue_1.6.2        purrr_0.3.5      
[21] pkgload_1.3.0     fastmap_1.1.0     compiler_4.2.1    processx_3.8.0   
[25] pkgbuild_1.3.1    sessioninfo_1.2.2 memoise_2.0.1     usethis_2.1.6    
> 

how to correct only tumor expression?

I think after MNN correction, both cell line expression and tumor sample expression value will be changed. is there any way to only correct tumor samples expression value, but keep the cell line expression value as reference, without any changes?