xchrom.tl.crosssamples_aucprc

xchrom.tl.crosssamples_aucprc(cell_embedding_ad: str | Path | AnnData, input_folder: str | Path = './test_data', output_path: str | Path = './eval_out', model_path: str | Path = './train_out/E1000best_model.h5', cellembed_raw: str = 'X_pca_harmony', save_pred: bool = False, scatter_plot: bool = False) → dict[source]

Evaluate the performance in cross-samples scenario, calculate auROC & auPRC for overall, per-cell and per-peak.

Parameters:

  • cell_embedding_ad (str or Path or anndata.AnnData) – Path to the cell embedding adata file. provide cell input embedding for XChrom model prediction.

  • input_folder (str or Path) – Path to the test data folder. Should generate by XChrom_preprocess.py. Must contain ‘splits.h5’, ‘ad.h5ad’, ‘all_seqs.h5’.

  • output_path (str or Path) – Path to the output folder.

  • model_path (str or Path) – Path to the trained model.

  • cellembed_raw (str) – Key of the raw cell input embedding in the cell embedding adata,to generate model input.

  • save_pred (bool) – Whether to save the prediction matrix with npy format.

  • scatter_plot (bool) – Whether to plot the UMAP of the test cells.

Returns:

metrics – Dictionary containing overall auROC, per-cell auROC, per-peak auROC, overall auPRC, per-cell auPRC, per-peak auPRC.

Return type:

dict

Examples

>>> import xchrom as xc
metrics1 = xc.tl.crosssamples_aucprc(
    cell_embedding_ad='./data/2_cross_samples/test_rna_harmony.h5ad',
    input_folder='./data/2_cross_samples/test_data',
    model_path='./data/2_cross_samples/train_out/E1000best_model.h5',
    output_path='./data/2_cross_samples/eval_out',
    cellembed_raw='X_pca_harmony',
    save_pred=True,
    scatter_plot=True
    )