Individual methods

Each notebook here drills into one annotation backend, calling its low-level entry point directly rather than going through ov.single.Annotation. Useful when you need to tune a specific method beyond what the unified Annotation.annotate(method=...) interface exposes, or when comparing how each backend behaves on the same input.

• Celltype auto annotation with SCSA
• Automatic cell type annotation with GPT/Other
• Celltype auto annotation with MetaTiME
• Using scMulan to annotate cell types in Heart, Lung, Liver, Bone marrow, Blood, Brain, and Thymus
• 1. load h5ad
• 2. transform original h5ad with uniformed genes (42117 genes)
• 3. process uniformed data (simply norm and log1p)
• 4. load scMulan
• 5. visualization
• Celltype annotation migration(mapping) with TOSICA

For the recommended, unified workflows that combine these methods, see:

• Reference-free annotation (Annotation API)

• Reference-based annotation (Annotation API)

• Consensus across methods (CellVote)