An experimental dense autoencoder that filters out SARS-CoV-2 sequences
Works best on sequences from Europe
This project requires conda. Get it here: https://anaconda.org/conda-forge/conda
To download the prerequisites run: conda env create -f conda.yaml
Drag your GISAID tar into ./data
Then run snakemake --cores all
The output is data/pruned.fasta that can be run through https://nextclade.org or any other tool