14-16 September 2026

Spatial omics technologies enable the simultaneous measurement of molecular profiles and spatial context within tissues, opening new avenues to study cellular organization, interactions, and function. However, analyzing spatial omics data requires integrating heterogeneous data types (transcripts, cells, images), performing robust quality control, and applying both classical spatial statistics and modern AI-based approaches.

This course provides an end-to-end, practical introduction to spatial omics data analysis, guiding participants from raw data preprocessing and quality control to advanced spatial analyses and AI-driven insights. Emphasis is placed on reproducible, SpatialData-centric workflows that unify multiple spatial modalities and scale to modern datasets.

Through a combination of lectures, hands-on tutorials, and guided discussions, participants will learn how to analyze spatial structure, cell–cell communication, and gene-level spatial patterns, and critically assess when and how AI-based methods add value in spatial biology.

This course is aimed at higher degree research students and early career researchers working with or with an interest in spatial omics and single-cell data. Some familiarity with Python  and scRNAseq is recommended. No prior experience with spatial omics data or AI methods is required.

Participants will learn how to:

● Perform robust preprocessing and quality control of spatial omics data

● Analyze spatial structure, cell–cell interactions, and gene-level patterns

● Understand when and how AI-based methods add value in spatial biology

● Apply modern tools within a coherent, reproducible SpatialData-centric workflow

Schedule: 4 hours per day (≈3.5 hours teaching + 30-minute break)
Teaching style: Lectures, hands-on tutorials, and guided discussions (via Zoom and Slack)

Day 1 — Preprocessing & Data Foundations - 2-6 PM Berlin time

Session 1: Introduction to SpatialData

• The SpatialData data model and ecosystem

• Unifying transcripts, cells, and images in a single framework

• Hands-on: loading, exploring, and inspecting SpatialData objects

Session 2: Quality Control in Spatial Omics

Diagnosing Segmentation Quality

• Common segmentation artifacts and failure modes

• Visual and quantitative QC strategies

• Hands-on: identifying problematic cells and regions

Segmentation Tools

• Overview of segmentation and preprocessing workflows

• Hands-on: running segmentation tools and comparing outputs

Session 3: SpatialData and Spatial Statistics

• Scalable spatial analysis concepts

• Building spatial neighbor graphs

• Hands-on: basic spatial statistics and visualizations

Day 2 — Core Spatial Analyses - 2-6 PM Berlin time

Session 4: Spatial Structure

Spatially Aware Clustering

• Concepts and evaluation metrics

• Hands-on: computing and interpreting spatial clusters

Niche Characterization

• Defining and quantifying cellular neighborhoods

• Hands-on: neighborhood enrichment and spatial niche visualization

Session 5: Cell–Cell Communication

• Spatial constraints in CCC inference

• Strengths and limitations of CCC methods

• Hands-on: spatially informed CCC analysis and visualization

Session 6: Gene-Level Spatial Analysis

Spatially Variable Genes

• Detecting spatial gene expression patterns

• Hands-on: SVG detection and interpretation

Imputation in Spatial Omics

• When imputation helps—and when it harms

• Hands-on: applying imputation and assessing downstream effects

Day 3 — AI-Driven Methods & Future Directions - 2-6 PM Berlin time

Session 7: Why AI in Spatial Omics?

• Motivations for AI-based models

• When data complexity justifies AI

• Interpretability, generalization, and trust

• Guided discussion on best practices and pitfalls

Session 8: AI Applications in Practice

Morphology and Multimodal Integration

• Extracting image-derived features

• Joint modeling of morphology and gene expression

• Hands-on / demo workflow

Deep Learning for Spatial Variability

• Modeling spatial expression patterns with deep learning

• Comparison with classical approaches

• Hands-on / demo and result interpretation

Closing Session: Synthesis & Outlook

• Summary of the full spatial analysis pipeline

• Common pitfalls and reproducibility best practices

• Open challenges and future directions in spatial omics

• Q&A and discussion of participant use cases

1 - Machine Learning for Multi-Omics Integration - ONLINE, 2-6 February

2 - Spatial Transcriptomics with Bioconductor - ONLINE, 9-13 March

3 - AI for Genomics - ONLINE, 7-9 April 

4 - Spatial Omics with R/Bioconductor - ONLINE, 18-20 May