Gene Panel Selection Workflow

To design a gene panel for scRNA-seq, use an end-to-end workflow combining dataset understanding, marker selection, seed-panel discovery, and biological completion. Provide your dataset and target panel size to generate a robust panel.

Selecting markers for spatial transcriptomics involves combining multiple methods like HVG, DE, Random Forest, and scGeneFit. This workflow evaluates panels using ARI, NMI, and SI benchmarking to ensure robust spatial analysis performance.

Yes, you can use scanpy and anndata for gene panel selection. The workflow depends on scanpy, anndata, pandas, and numpy to process single-cell datasets and execute the marker selection and panel completion pipeline.

ARI benchmarking evaluates single-cell gene panels by measuring clustering similarity against the original dataset. The workflow uses ARI, NMI, and SI metrics to compare multiple marker-selection methods and verify panel robustness across downsampling scenarios.

Automated gene panel design without biological curation may miss critical context-specific markers. This workflow addresses limitations by performing exhaustive literature-based marker lookup and category-balanced panel assembly to ensure biological relevance after computational seed-panel discovery.