Skip to content

Latest commit

 

History

History
112 lines (85 loc) · 4.98 KB

README.md

File metadata and controls

112 lines (85 loc) · 4.98 KB
Raw

Enterovirus A71 Nextstrain Analysis

This repository provides a comprehensive Nextstrain analysis of Enterovirus A71. You can choose to perform either a VP1 run (>=600 base pairs) or a whole genome run (>=6400 base pairs).

For those unfamiliar with Nextstrain or needing installation guidance, please refer to the Nextstrain documentation.

Enhancing the Analysis

This analysis would benefit from additional metadata, such as patient age, spatial data, and clinical outcomes. If you have relevant data and are willing to share, please contact us.

The data for this analysis is available from NCBI Virus. Instructions for downloading sequences are provided under Sequences.

Repository Organization

This repository includes the following directories and files:

  • scripts: Custom Python scripts called by the snakefile.
  • snakefile: The entire computational pipeline, managed using Snakemake. Snakemake documentation can be found here.
  • ingest: Contains Python scripts and the snakefile for automatic downloading of EV-A71 sequences and metadata.
  • vp1: Sequences and configuration files for the VP1 run.
  • whole_genome: Sequences and configuration files for the whole genome run.

Configuration Files

The config, vp1/config, and whole_genome/config directories contain necessary configuration files:

  • colors.tsv: Color scheme
  • geo_regions.tsv: Geographical locations
  • lat_longs.tsv: Latitude data
  • dropped_strains.txt: Dropped strains
  • clades_genome.tsv: Virus clade assignments
  • reference_sequence.gb: Reference sequence
  • auspice_config.json: Auspice configuration file

The reference sequence used is BrCr, accession number U22521, sampled in 1970.

Quickstart

Setup

Nextstrain Environment

Install the Nextstrain environment by following these instructions.

Running a Build

Activate the Nextstrain environment:

conda activate nextstrain

To perform a build, run:

snakemake --cores 1

For specific builds:

  • VP1 build: 
    snakemake auspice/ev_a71_vp1.json --cores 1
  • Whole genome build: 
    snakemake auspice/ev_a71_whole_genome.json --cores 1

First steps

To run the ingest, you will need some specific reference files, such as a reference.fasta or annotation.gff3 file.

  1. In the config file: check that the taxid is correct

  2. To get these reference files you have to run the script generate_from_genbank.py manually.

    python3 ingest/generate_from_genbank.py --reference "U22521.1" --output-dir ingest/data/references/

    • While the code runs, you will need to specify a few things:

      1. [0]
      2. [product] or [leave empty for manual choice] if you want to specify the proteins
      3. [2].

    • It will create the files in the subdirectory data/references.

    • These files will be used by the ingest snakefile.

  3. Check that the attributes in data/references/pathogen.json are up to date.

  4. Run the `ingest' snakefile (either manually or using the main snakefile).

    • Depending on your system you may need to run chmod +x ./vendored/*;chmod +x ./bin/* first.

  5. Run the main snakefile.

Visualizing the Build

To visualize the build, use Auspice:

auspice view --datasetDir auspice

To run two visualizations simultaneously, you may need to set the port:

export PORT=4001

Sequences

Sequences can be downloaded manually or automatically.

  1. Manual Download: Visit NCBI Virus, search for EV-A71 or Taxid 39054, and download the sequences.
  2. Automated Download: The ingest functionality, included in the main snakefile, handles automatic downloading.

The ingest pipeline is based on the Nextstrain RSV ingest workflow. Running the ingest pipeline produces data/metadata.tsv and data/sequences.fasta.

Updating Vendored Scripts

This repository uses git subrepo to manage copies of ingest scripts in ingest/vendored. To pull new changes from the central ingest repository, first install git subrepo and then follow the instructions in ingest/vendored/README.md.

Feedback

For questions or comments, contact me via GitHub or [email protected]

To Do:

  • Use the latest version of Auspice (v25)
  • Overwrite NCBI virus metadata with "corrected" collection dates
  • Replace parse_date with augur curate
  • Provide a way to create and use "local" accession numbers for sequences not on Genbank yet.