reclustering.yaml Reference

This configuration file corresponds to --option=reclustering and is used for secondary subdivision of target populations and subcluster marker identification.

Parameter	Default	Description
option	reclustering	Fixed identifier for the analysis stage
results_folder	results	Root directory for analysis outputs
sample_list	sample.txt	Path to the sample list file
channel	single_analysis	Analysis channel
run_type	visium	Platform type
recluster_resolution	0.8	Resolution used for reclustering
recluster_n_top_genes	2000	Number of highly variable genes
recluster_neighbors	15	Number of neighbors used for the neighbor graph
recluster_n_pcs	30	Number of PCA dimensions used in reclustering
recluster_marker_method	wilcoxon	Statistical method used for subcluster marker detection
recluster_min_pct	0.1	Minimum positive fraction threshold for marker detection
recluster_logfc_threshold	0.25	Minimum logFC threshold for marker detection

Tuning suggestions

1. Tune recluster_resolution before recluster_n_pcs so the effect of each parameter is easier to interpret. Because reclustering usually targets a smaller dataset, modest values are often sufficient.

2. recluster_min_pct and recluster_logfc_threshold jointly determine how stringent marker detection will be.