Implement reference mapping components #138
Conversation
|
The repository is ready for review. I removed unnecessary details and fixed bugs. Local tests are running fine, but the test pipeline will fail due to the lack of test files. @rcannood , can you please run the script workflows/resources_test_scripts/hlca_reference_model.sh and put the resulting files on AWS so that test script could use them? |
src/integrate/scarches/script.py
Outdated
|
|
||
| with tempfile.TemporaryDirectory() as directory: | ||
| model_path = _download_HLCA_reference_model(directory) | ||
| vae_query, adata_query = map_to_existing_reference(adata_query, model_path=model_path) |
There was a problem hiding this comment.
True. One way to do this is to make the reference type file and set the default to the url (https://zenodo.org/record/6337966/files/HLCA_reference_model.zip). However, this also means that if we want to pass another reference, this also needs to be a path/url. FYI: if we upload them to s3, s3:// urls are also supported. I think just using the zenodo URL is a better solution.
Correct me if im wrong @rcannood
|
LGTM! Very nice job @VladimirShitov, thanks! |
integrate/scarchescomponent to map query data to the reference (currently only supports HLCA)integrate/scvi_toolscomponent to build reference and map query data on itintegrate/scvi_toolsintegrate/scvi_toolsintegrate/scvi_tools