Note
This workshop is under development. Dates, registration and materials will be added here as they are finalised.
A two-day, hands-on workshop on analysing spatial biology data. Day 1 uses QuPath to explore, segment and classify cells interactively. Day 2 rebuilds the same analysis in Python, then goes further with transcripts, segmentation quality control and spatial statistics.
Both days use the same public tissue section, which has multiplex immunofluorescence (protein), H&E and in situ transcript data.
By the end of the workshop you should be able to:
- Explain how multiplex IF, H&E and in situ transcript data relate on one section, and what each can and can't tell you.
- Segment nuclei and cells, and measure per-cell marker intensities and transcript counts.
- Classify cells by marker thresholds (QuPath) and by unsupervised clustering (Python).
- Describe spatial organisation: neighbourhoods, cell–cell co-location, tissue niches.
- Move results between QuPath and Python, and judge which tool suits which question.
The workshop will not analyse your own data. Its aim is to give you enough grounding to choose the right tools for your data, and to ask the right questions when evaluating new ones.
Registration is by application and is not yet open. The application form asks about your research and what you hope to get from the workshop.
TBC
Before the workshop:
- Install QuPath and the Python environment, following the installation instructions. The Python environment works on Windows, Linux and macOS (Apple Silicon).
- From the repository folder, check the Python environment with
pixi run check-full. This also downloads the cell segmentation model, so run it with an internet connection and preferably off any VPN. - Download the workshop data (link to follow).
| Session | Topic |
|---|---|
| 1 | Digital images and an introduction to spatial biology |
| 2 | H&E workflow: tissue detection, region classification, cell detection |
| 3 | Multiplex IF workflow: cell segmentation with InstanSeg, cell classification, phenotypes |
| 4 | Spatial analysis in QuPath, and exporting results for Day 2 |
| Session | Topic |
|---|---|
| 5 | Jupyter notebooks, the SpatialData data model, loading the data |
| 6 | Cell segmentation and quality control |
| 7 | Cell types: clustering and comparison with Day 1 |
| 8 | Spatial statistics and niches, and where to go next |
TBC
TBC
- CellSurvey: a pipeline that runs the Day 2 analysis end to end, built on Sopa.
- Introduction to image analysis: our general image-analysis workshop.
TBC