OmicVerse Single-Cell LDA Topic Clustering

Goal

Turn the notebook's MIRA-backed topic modeling section into a reusable job: fit an LDA topic model on count-like single-cell data, predict topic usage, and optionally derive hard cluster labels or RFC-based labels.

Quick Workflow

1. Inspect whether the data still has a count layer and a highly-variable-feature flag.
2. Choose feature_type, learning_rate, and whether to run ondisk=False or ondisk=True.
3. Construct LDA_topic(...), then decide how many topics to fit.
4. Run predicted(num_topics=...) to create LDA_cluster.
5. If the user wants harder classification on an embedding, run get_results_rfc(...) with explicit use_rep and LDA_threshold.

Interface Summary