cellextract: image-to-feature nextflow pipeline

This pipeline processes microscopy images to single-cell feature matrices. It has found application in our work on phenotypes induced by microRNAs and was developed primarily for high-content (Cell Painting) data in mind.

Features

• Segmentation with Cellpose-2S, which detects single- and multi-nucleated cells
• Image QC feature extraction (thanks to hooks into CellProfiler)
• CellProfiler-like feature extraction

Running the Workflows

1. Adapt the example_nextflow.config to your needs (see below)
2. Run the pipeline nexflow run (possibly with -profile and -c options to point to your nextflow profiles and configs)

For example, you may wish to use singularity/docker so you do not have to manage dependencies yourself. To do so, make sure you have singularity installed, then use nextflow run -c singularity.

Config

Steps to adapt the config:

1. Copy the config to one that nextflow will detect automatically: cp example_nextflow.config nextflow.config
2. Edit it using your favorite text editor, e.g. nano nextflow.config