Converting Leica Versa SCN files to OME-TIFF

The following scripts help in the conversion of SCN files from Leica Versa slide scanners to OME-TIFF images that are tiled, pyramidal and contain a single scene. It also removes additional metadata from the OME-XML (acquisition data and strutured annotations) to ensure that files are de-identified.

The first series in a SCN file is (usually) the macro image. Conversion of this is skipped. All other scenes are output to individual OME-TIFF files.

JPEG compression us used but this could be changed in twosteps.sh

Requirements

bioformats2raw>=0.5 - this is in the ome Anaconda channel but not for arm-64. If on a Apple Silicon Mac you may need to specify a x86 conda env or install from Github
raw2ometiff>=0.4 - this is not on the ome Anaconda channel yet so will need to be installed from Github and added to your path.
Python >=3.7
ome_types - Installable on pip

Usage

If input.scn contains 3 series of which the first is a macro image

bash convert_scn.sh input.scn

Outputs:

inputs-scene1.ome.tiff
inputs-scene2.ome.tiff

Docker usage

A docker container with prereqiisites on Dockerhub (adamjtaylor/scn2ometiff) and quay.io (quay.io/adamjtaylor/scn2ometiff)

Use as follows, mounting the directory with the input images into the container

docker run -it --rm \
  -v <local-dir>:/data adamjtaylor/scn2ometiff \
  /bin/bash -c "/convert_scn.sh /data/<image-to-convert>.scn"

There may be warnings about container architecure and OpenCV

Nextflow usage

⚠️ In development ⚠️ May not be functional

There is also a Nextflow pipeline for reproducible and contanarised conversion of scn files

To convert all .scn files in the input directory and output into the output directory run:

nextflow run ncihtan/scn2ometiff -r nf --input input/*.scn --output output