Skip to content

What is multi-catalog.yaml?

Multi-Sample Catalog YAML File (multi-catalog.yaml): the top-level configuration file for multi-sample spatial transcriptomics experiments and shared spatial factor analyses in CartoScope. It serves as a master index that links multiple individual sample datasets into a single harmonized session.

What does multi-catalog.yaml define?

  • analysis_type: multi-sample: Identifies the catalog as a multi-sample experiment rather than a single slide.
  • name / title: The session identifier or study name for the multi-sample dataset collection.
  • samples: A dictionary mapping sample IDs/names to individual sample catalog files (e.g. ./sample_1/catalog.yaml or remote HTTPS/S3 URLs).
  • factors: Shared FICTURE spatial factor set definitions computed across all participating samples.
  • countsUrl / sge_counts: Path to the unified SGE expression matrix used for pseudobulk calculations, sample label grouping, Chi-squared testing, and DESeq2 differential expression analysis.

What does a typical multi-catalog.yaml look like?

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
id: mouse_brain_replicates_2026
name: Mouse Brain Cortex Replicates Study
analysis_type: multi-sample

samples:
  Control_Rep1: ./control_1/catalog.yaml
  Control_Rep2: ./control_2/catalog.yaml
  Treated_Rep1: ./treated_1/catalog.yaml
  Treated_Rep2: ./treated_2/catalog.yaml

factors:
  - id: ficture_24_factors
    title: Shared 24-Factor FICTURE LDA Model
    path: ./shared_factors/ficture_24.parquet

countsUrl: ./matrices/multi_sample_sge.parquet

Why use multi-catalog.yaml?

  • Automated Session Setup: Automatically resolves sample metadata, thumbnails, and grid layouts upon loading.
  • Shared Spatial Factors: Aligns shared FICTURE spatial factor signatures across all participating slides.
  • Unified Statistical Analysis: Enables cross-sample pseudobulk calculations, sample label grouping, Chi-squared testing, and DESeq2 differential expression.

Related Pages

Concepts

Manual Pages