Hello,Thank you very much for your excellent work. This resource has been extremely helpful for human intestinal single-cell research.
I encountered a possible file mismatch when using the epithelial large intestine atlas resources. I mapped my query data using the following scANVI model:
LI_AP_epi_scanvi_fine_annot.updated20240131
This generated X_scANVI embeddings for the query data. However, when I tried to combine these embeddings with the reference object:
29_Extended_Pan-GI_atlas_Epi_LI_AP_20241119.h5ad
the embeddings did not appear to be in the same latent space.
In addition, I noticed two potential inconsistencies:
The KNN model
knn_label_transfer_model_healthy_reference_Epi_LI_AP_20241005.pkl
appears to correspond to SI_AP rather than LI_AP, as its size seems to match the SI_AP model.
The number of cells in
Epi_LI_AP_obs.csv
does not match the number of healthy_reference cells in
29_Extended_Pan-GI_atlas_Epi_LI_AP_20241119.h5ad.
Could you please help check whether the uploaded LI_AP scANVI model, KNN model, and reference metadata are correctly matched? If there was a file upload mismatch, would it be possible to update the corresponding files?
Thank you again for providing such a valuable resource. I would greatly appreciate your help.
Hello,Thank you very much for your excellent work. This resource has been extremely helpful for human intestinal single-cell research.
I encountered a possible file mismatch when using the epithelial large intestine atlas resources. I mapped my query data using the following scANVI model:
LI_AP_epi_scanvi_fine_annot.updated20240131
This generated X_scANVI embeddings for the query data. However, when I tried to combine these embeddings with the reference object:
29_Extended_Pan-GI_atlas_Epi_LI_AP_20241119.h5ad
the embeddings did not appear to be in the same latent space.
In addition, I noticed two potential inconsistencies:
The KNN model
knn_label_transfer_model_healthy_reference_Epi_LI_AP_20241005.pkl
appears to correspond to SI_AP rather than LI_AP, as its size seems to match the SI_AP model.
The number of cells in
Epi_LI_AP_obs.csv
does not match the number of healthy_reference cells in
29_Extended_Pan-GI_atlas_Epi_LI_AP_20241119.h5ad.
Could you please help check whether the uploaded LI_AP scANVI model, KNN model, and reference metadata are correctly matched? If there was a file upload mismatch, would it be possible to update the corresponding files?
Thank you again for providing such a valuable resource. I would greatly appreciate your help.