Input Data Summary
------------------
File: data/kretzler_kidney.h5ad
Dimensions (Cells x Genes): 2000 x 36020

Structure:
- Data contains scRNA-seq counts.
- Genes (variables) are identified by symbols in var.index.
- 'obs' dataframe contains author-provided annotations.

Usage for this task:
The author annotations found in the 'obs' dataframe (using column 'cell_type') are extracted to serve as the ground truth. These are compared against the SCimilarity model's predictions to assess the model's accuracy on the kidney dataset and generate the concordance heatmap (Figure 3d).