nextstrain.org/lassa dmsa_pred build

For those unfamiliar with Nextstrain, we recommend checking out the Nextstrain documentation before reading further.

This repository is forked from nextstrain.org/lassa and modified to include the analysis of deep mutational scanning data. The analysis is performed using the dmsa_pred package. This repo is currently set up to run the data as presented in the manuscript Deep mutational scanning reveals functional constraints and antigenic variability of Lassa virus glycoprotein complex.

To run the pipeline, install snakemake following their documentation, This pipeline has been tested with versions >= 8.4.11

then clone this repository and the submodules

git clone https://github.com/matsengrp/lassa-dmsa.git --recurse-submodules

Run the pipeline with the following command

snakemake --use-conda --cores 2 --configfile config/config.yaml

Each rule currently uses the environment specified in my_profiles/dmsa_pred/dmsa_env.yaml. The configuration file specified above provides paths to the escape data within the LASV_Josiah_GP_DMS submodule. The pipeline will output a JSON file(s) under the "auspice/" directory that can be visualized with auspice software. CSV's of phenotype predictions can be found under "results/dmsa-phenotype/"

Genbank accessions

To download the accessions, go to NCBI Virus and click Search by virus. In the Search by virus name or taxonomy box, enter Mammarenavirus lassaense, taxid:3052310 and hit enter. Then click the Download option, select Accession List and Nucleotide options and hit Next. On the next page, select Download All Records and hit Next. On the next page, select Accession with version and click Download. Sequences are downloaded from the list of accessions because more information is extracted from the genbank file during the download process. The current accession list was downloaded on August 10, 2023.

Configuration

Configuration takes place within the Snakefile and the config/config.yaml files. The Snakefile can be read top-to-bottom, each rule specifies its file inputs and output and also its parameters. There is little redirection and each rule should be able to be reasoned with on its own. The config/config.yaml is important for configuring the DMSA phenotype prediction component of this build (e.g., paths to the deep mutational scanning directories for escape data).