Revert "temporarily remove cell cell comm"

[scottgigante-immunai]

This reverts commit 1068e5f.

Cell cell communication was causing benchmarks to fail: https://github.com/openproblems-bio/openproblems/actions/runs/3136011677/jobs/5095990766

We need to identify the problem before we can merge it back in.

[codecov]

Codecov Report

Base: 86.11% // Head: 94.62% // Increases project coverage by +8.50% 🎉

Coverage data is based on head (ca80b97) compared to base (f9ecd71).
Patch has no changes to coverable lines.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #598      +/-   ##
==========================================
+ Coverage   86.11%   94.62%   +8.50%     
==========================================
  Files         135      135              
  Lines        3378     3440      +62     
  Branches      176      176              
==========================================
+ Hits         2909     3255     +346     
+ Misses        412      124     -288     
- Partials       57       61       +4     

Flag	Coverage Δ
unittests	94.62% <ø> (+8.50%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
openproblems/data/immune_cells.py	100.00% <0.00%> (ø)
.../_batch_integration/batch_integration_embed/api.py	100.00% <0.00%> (ø)
.../_batch_integration/batch_integration_graph/api.py	100.00% <0.00%> (ø)
...gration/batch_integration_graph/datasets/immune.py	100.00% <0.00%> (ø)
...gration/batch_integration_graph/methods/fastmnn.py	100.00% <0.00%> (ø)
...ation/batch_integration_graph/datasets/pancreas.py	100.00% <0.00%> (ø)
...lems/tasks/_cell_cell_communication/_common/api.py	91.57% <0.00%> (+91.57%)	⬆️
openproblems/data/tnbc_wu2021.py	100.00% <0.00%> (+100.00%)	⬆️
openproblems/data/allen_brain_atlas.py	100.00% <0.00%> (+100.00%)	⬆️
...ms/tasks/_cell_cell_communication/_common/utils.py	100.00% <0.00%> (+100.00%)	⬆️
... and 8 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[dbdimitrov]

Ooh, debug this in the next couple of days!!

[scottgigante-immunai]

Error code from the nextflow log looks like it's to do with the location of the omnipathr config file.

WARNING:rpy2.rinterface_lib.callbacks:R[write to console]: In addition:
WARNING:rpy2.rinterface_lib.callbacks:R[write to console]: Warning message:
WARNING:rpy2.rinterface_lib.callbacks:R[write to console]: In file(file, "w", encoding = fileEncoding) :
WARNING:rpy2.rinterface_lib.callbacks:R[write to console]: 
WARNING:rpy2.rinterface_lib.callbacks:R[write to console]: cannot open file '/home/sagemaker-user/.config/OmnipathR/omnipathr.yml': Permission denied
Traceback (most recent call last):
  File "/usr/local/bin/openproblems-cli", line 33, in <module>
    sys.exit(load_entry_point('openproblems', 'console_scripts', 'openproblems-cli')())
  File "/usr/src/singlecellopenproblems/openproblems/api/main.py", line 45, in main
    output = _main(args)
  File "/usr/src/singlecellopenproblems/openproblems/api/main.py", line 38, in _main
    return SUBCOMMANDS[args.subcommand].main(args)
  File "/usr/src/singlecellopenproblems/openproblems/api/load.py", line 12, in main
    adata = load_dataset(args.task, args.name, args.test)
  File "/usr/src/singlecellopenproblems/openproblems/api/load.py", line 7, in load_dataset
    return fun(test=test)
  File "/usr/src/singlecellopenproblems/openproblems/tools/decorators.py", line 155, in apply_func
    adata = func(*args, **kwargs)
  File "/usr/src/singlecellopenproblems/openproblems/tasks/_cell_cell_communication/cell_cell_communication_ligand_target/datasets/tnbc_wu2021.py", line 24, in tnbc_data
    adata.uns["ligand_receptor_resource"] = ligand_receptor_resource(
  File "<decorator-gen-33>", line 2, in __call__
  File "/usr/local/lib/python3.8/site-packages/scprep/utils.py", line 80, in _with_pkg
    return fun(*args, **kwargs)
  File "/usr/local/lib/python3.8/site-packages/scprep/run/r_function.py", line 138, in __call__
    robject = self.function(*args, **kwargs)
  File "/usr/local/lib/python3.8/site-packages/rpy2/robjects/functions.py", line 198, in __call__
    return (super(SignatureTranslatedFunction, self)
  File "/usr/local/lib/python3.8/site-packages/rpy2/robjects/functions.py", line 125, in __call__
    res = super(Function, self).__call__(*new_args, **new_kwargs)
  File "/usr/local/lib/python3.8/site-packages/rpy2/rinterface_lib/conversion.py", line 45, in _
    cdata = function(*args, **kwargs)
  File "/usr/local/lib/python3.8/site-packages/rpy2/rinterface.py", line 677, in __call__
    raise embedded.RRuntimeError(_rinterface._geterrmessage())
rpy2.rinterface_lib.embedded.RRuntimeError: Error: .onLoad failed in loadNamespace() for 'OmnipathR', details:
  call: file(file, "w", encoding = fileEncoding)
  error: cannot open the connection

<error/rlang_error>
Error:
! .onLoad failed in loadNamespace() for 'OmnipathR', details: call:
  file(file, "w", encoding = fileEncoding) error: cannot open the connection
---
Backtrace:
    ▆
 1. ├─`<fn>`(9606L)
 2. └─base::loadNamespace(x)
 3.   ├─base::namespaceImportFrom(...)
 4.   │ └─base::asNamespace(ns)
 5.   └─base::loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]])
 6.     └─base (local) runHook(".onLoad", env, package.lib, package)`, size: 3107 (max: 255)

[LuckyMD]

Is Omnipath ever explicitly called in your code, or is it wrapped by LIANA? This seems to error out in a load_dataset function, but the tnbc_wu dataset never calls omnipath explicitly as far as I can tell.

[dbdimitrov]

OK. This is odd.

OmniPath is not actually used (I thought it might be wrongly calling liana::generate_homologs), but it is indeed loaded implicitly (wrapped in LIANA).

Loading OmniPath then makes it initiate it's own .cache and .configs, and this is where/why the exception is thrown. I was able to reproduce it by removing the permissions from my OmnipathR/omnipathr.yml file, meaning that the sagemaker-user run does not have access to it's own home directory? Is there a way to allow it to save small files (it won't be anything larger than 5MB)? This might be the easiest solution.

The next in-line solution would be to enable OmniPath to save in the current directory, or alternatively to just have a temporary omnipath.env created each time it's run (thus not writing anything at all).

Third solution (most work solution), save everything as files and I can move OmniPath from imported to suggested in LIANA.

[scottgigante-immunai]

I just opened up the docker image and sagemaker-user does indeed have access to its home directory, however, it does not have access to ~/.config. This should be an easy fix.