Spatial transcriptomics data analysis

This page contains an overview of all the material you need to follow the spatial transcriptomics data analysis training.

Software

• VSCode with Python and Jupyter plugin: check the topic on Jupyter in our preparatory trainings
• miniconda3, git and the .yml file for creating the conda environment: check the topic on conda in our preparatory trainings
• Qupath (0.6.0 or 0.5.1)
• Fiji and Bigwarp

Data

• Backup data for BigWarp (2.5 Gb)

Notebooks

Open a git bash and run: git clone https://github.com/vibspatial/targeted_transcriptomics_training.git

Archive

Below you can find the slides archive of the theoretical spatial omics data analysis in 2023 and 2024.

• Introduction (Julien Mortier – Spatial Catalyst)
• Targeted transcriptomics and the SPArrOW analysis pipeline (Lotte Pollaris – Yvan Saeys lab)
• Visualization tools (Arne Defauw – Spatial Catalyst)
• Analysis of untargeted transcriptomics data (Frank Vernaillen – Spatial Catalyst)
• Downstream spatial analysis (Julien Mortier – Spatial Catalyst)
• Outro (Julien Mortier – Spatial Catalyst)

Why you should process your data

Instead of using the processed data that is offered by the vendors you should do the preprocessing of the data yourself because the vendors make mistakes:

• the vendors sometimes miss cells during the cell segmentation
• some vendors expand the cells beyond the nucleus after cell segmentation leading to “cells” that are actually a mix of multiple real cells
• some vendors bin the data leading to bins that are a mix of multiple real cells

For analysis of spatial transcriptomics data we will use the sparrow pipeline that was created by Lotte Pollaris from the VIB Saeys Lab. She presents het work in all the videos with theoretical background in this lesson.