siqc: Generate standalone interactive QC reports from scientific data

Generate interactive QC reports from scientific data with progressive loading and spatial visualization. Perfect for single-cell genomics, spatial transcriptomics, and other large-scale biological datasets.

Quick start: npm install -g siqc → siqc --help

Features

• 🚀 Progressive Loading: Large datasets (millions of cells) load incrementally without blocking the UI
• 📊 Interactive Visualizations: Histograms, scatter plots, heatmaps, and spatial plots with Plotly.js
• 🗜️ Efficient Compression: Binary data format with gzip compression for optimal file sizes
• 📄 Standalone Reports: Single HTML file with no external dependencies
• 🎯 Spatial Analysis: Advanced spatial plotting with faceting and optimization
• ⚡ Performance Optimized: Web Workers, typed arrays, and multi-scale rendering
• 🛠️ Easy CLI: Simple command-line interface for quick report generation

Installation

From npm (Recommended)

# Install globally from npm
npm install -g siqc

From Git Repository

# Install globally from git repository (development version)
npm install -g git+https://github.com/openpipelines-bio/siqc.git

Local Development

git clone https://github.com/openpipelines-bio/siqc.git
cd siqc
npm install
# In development, CLI is available via `npm run cli` instead of global `siqc`

Quick Start

1. Generate Example Data

# Generate single-cell test data
siqc generate-test-data --type sc --output ./example-data

# Generate spatial test data  
siqc generate-test-data --type xenium --output ./spatial-data

2. Create Your First Report

# Render report from test data
siqc render --data ./example-data/data.json --structure ./example-data/structure.json --output ./my-report.html

# Open the report in your browser
open my-report.html

3. Use Your Own Data

Prepare your data in the required format (see Data Format) and render:

siqc render --data ./my-qc-data/data.json --structure ./my-qc-data/structure.json --output ./my-qc-report.html

CLI Reference

Commands

generate-test-data

Generate example datasets for testing and development.

siqc generate-test-data --type <type> --output <directory>

Options:

• --type, -t: Dataset type (sc, sc_large, xenium, xenium_large, cosmx, cosmx_large, visium, visium_large)
• --output, -o: Output directory for generated files
• --verbose: Enable verbose logging

Examples:

siqc generate-test-data --type sc --output ./test-data
siqc generate-test-data --type cosmx --output ./cosmx-data
siqc generate-test-data --type visium --output ./visium-data
siqc generate-test-data --type xenium_large --output ./large-spatial --verbose

render

Generate QC report from existing data.

siqc render --data <file> --structure <file> --output <file> [options]

Options:

• --data, -d: Path to data JSON file (required)
• --structure, -s: Path to structure JSON file (required)
• --output, -o: Output HTML file path (required)
• --payload, -p: Path to binary payload file (auto-generated if not specified)
• --auto-generate: Auto-generate payload file if not specified (default: true)
• --verbose: Enable verbose logging

Examples:

siqc render --data ./data.json --structure ./structure.json --output ./report.html
siqc render --data ./data.json --structure ./structure.json --output ./report.html --payload ./cached-payload.bin

compress

Compress data to binary payload format for faster loading.

siqc compress --data <file> --structure <file> --output <file>

Options:

• --data, -d: Path to input data JSON file (required)
• --structure, -s: Path to structure JSON file (required)
• --output, -o: Output binary payload file path (required)
• --verbose: Enable verbose logging

Examples:

siqc compress --data data.json --structure structure.json --output payload.bin

Data Format

data.json

Your data should be in columnar format for optimal performance:

{
  "cell_rna_stats": {
    "columns": [
      {
        "name": "sample_id",
        "dtype": "categorical",
        "data": [0, 0, 1, 1, 2],
        "categories": ["sample_A", "sample_B", "sample_C"]
      },
      {
        "name": "total_counts",
        "dtype": "integer",
        "data": [1200, 1500, 800, 2000, 1100]
      },
      {
        "name": "fraction_mitochondrial",
        "dtype": "numeric", 
        "data": [0.05, 0.08, 0.12, 0.03, 0.07]
      }
    ]
  },
  "spatial_coords": {
    "columns": [
      {
        "name": "x_coord",
        "dtype": "numeric",
        "data": [100.5, 200.3, 300.1, 150.7, 250.9]
      },
      {
        "name": "y_coord", 
        "dtype": "numeric",
        "data": [50.2, 75.8, 90.3, 120.5, 80.1]
      }
    ]
  }
}

structure.json

Define the report structure and categories:

{
  "title": "My QC Report",
  "categories": [
    {
      "key": "cell_rna_stats",
      "name": "Cell RNA Statistics",
      "additionalAxes": false,
      "defaultFilters": []
    },
    {
      "key": "spatial_coords", 
      "name": "Spatial Coordinates",
      "additionalAxes": true,
      "defaultFilters": []
    }
  ]
}

Data Types:

• categorical: Integer indices with category labels
• integer: Whole numbers
• numeric: Floating-point numbers
• boolean: True/false values