covid-spike-classification is a script to call interesting SARS-CoV-2 spike protein mutations from Sanger sequencing to support the Danish COVID-19 monitoring efforts.

Using Sanger-sequenced RT-PCR product of the spike protein, this tool should pick up all relevant mutations currently tracked (see covid_spike_classification/core.py for the full list of tracked mutations) and give a table with one row per sample and a yes/no/failed column per tracked mutation.

This workflow is built and maintained at https://github.com/kblin/covid-spike-classification

If you found this tool useful, please cite https://www.medrxiv.org/content/10.1101/2021.03.27.21252266v1

covid-spike-classification is distributed via this git repository, pypi or bioconda.

Installing via bioconda is the fastest way to get up and running:

conda create -n csc -c conda-forge -c bioconda covid-spike-classification
conda activate csc

When installing via git or pypi, you first need to install the external binary dependencies.

covid-spike-classification depends on three excellent tools to do most of the work:

• tracy (versions 0.5.3 & 0.5.7 tested)
• bowtie2 (version 2.4.2 tested)
• samtools (versions 1.10 & 1.11 tested)

If you have conda installed, the easiest way to get started is to just install these via calling

git clone https://github.com/kblin/covid-spike-classification.git
cd covid-spike-classification
conda env create -n csc -f environment.yml
conda activate csc
pip install .

While not technically an installation method, covid-spike-classification is also shipped as an OCI container. To use it, you ideally run the container from a workflow management system like Snakemake or Nextflow that will take care of mounting filesystems into the container for you.

The OCI container image is available from the Docker Hub kblin/covid-spike-classification repository.

You also need to generate the samtools and bowtie2 indices for your reference genome. We ship a copy of NC_045512 and a script to generate these indices:

conda activate csc
cd ref
./build_indices.sh
cd ..

Assuming you used above instructions to install via conda, you can run the tool like this:

conda activate csc
covid-spike-classification --reference /path/to/your/reference.fasta --outdir /path/to/result/dir /path/to/sanger/reads/dir_or.zip

Notably, you can provide the input either as a ZIP file or as a directory, as long as they directly contain the ab1 files you want to run the analysis on.

See also the --help output for more detailed usage information.

All code is available under the Apache License version 2, see the LICENSE file for details.