Inconsistent preprocessing for scPRINT

[mumichae]

Describe the bug
Feature selection step seems to be missing.

task_batch_integration/src/methods/scprint/script.py

Lines 47 to 57 in 4e9ebb6

	print("\n>>> Preprocessing data...", flush=True)
	preprocessor = Preprocessor(
	min_valid_genes_id=min(0.9 * adata.n_vars, 10000), # 90% of features up to 10,000
	# Turn off cell filtering to return results for all cells
	filter_cell_by_counts=False,
	min_nnz_genes=False,
	do_postp=False,
	# Skip ontology checks
	skip_validate=True,
	)
	adata = preprocessor(adata)

Expected behavior
Feature selection should be handled via command line argument

task_batch_integration/src/methods/scvi/config.vsh.yaml

Lines 18 to 21 in 4e9ebb6

	- name: --n_hvg
	type: integer
	default: 2000
	description: Number of highly variable genes to use.

task_batch_integration/src/methods/scvi/script.py

Lines 32 to 35 in 4e9ebb6

	if par["n_hvg"]:
	print(f"Select top {par['n_hvg']} high variable genes", flush=True)
	idx = adata.var["hvg_score"].to_numpy().argsort()[::-1][:par["n_hvg"]]
	adata = adata[:, idx].copy()

Additional context
If full feature matrix for the model is desired by default, this should be achieved by adjusting the --n_hvg parameter in, or adding new variant here:

task_batch_integration/src/methods/scprint/config.vsh.yaml

Lines 34 to 40 in 4e9ebb6

	variants:
	scprint_large:
	model_name: "large"
	scprint_medium:
	model_name: "v2-medium"
	scprint_small:
	model_name: "small"

[jkobject]

scPRINT needs the full anndata to work well as of now. It doesn't do feature selection and only selects the genes that are expressed in each specific cells.

Let me me know if you have other questions :)