DFAST_VRL

DFAST_VRL is a viral genome annotation and data submission tool to DDBJ.
It internally uses NCBI-VADR and can generate data submission files to DDBJ.

Pipeline

1. Preprocessing
   Query genome is mapped to the reference to calculate the coverage against the reference. Runs of leading/trailing Ns are trimmed.
   (Optional) Unmapped regions are removed, and mapped regions are scaffoled (ordered and concatenated with runs of Ns).
2. Annotation
   DFAST_VRL internally uses NCBI-VADR to annotate bilogical features (CDS, mature peptide, ribosomal slippage, stem loop).
3. Variant detection
   Based on the pairwise alignment with the reference genome using MAFFT
4. Output
   GenBank format, VCF format, DDBJ data submission format

Dependencies

• NCBI VADR
• NCBI-BLAST+ (included in the VADR package)
• snpEff
• MAFFT
• Biopython

Since DFAST_VRL depends on many external binaries and modules, it is recommended to use Docker/Singularity container as described below.

Usage

dfast_vrl -i input_genome.fasta [-o output_directory] [-m metadata_file.txt]

Example:

./dfast_vrl -i examples/LC570964-6.draft.contigs.fa -m examples/metadata.txt

The input file must be a single-FASTA file consisting of one complete genome or a multi-FASTA file consisting of multiple sequences from a draft genome assebmly.

For low-quality genome, use --as_misc to annotatte partial/ambiguous CDS as misc_feature

./dfast_vrl -i examples/LC570964-6.draft.contigs.fa -m examples/metadata.txt --as_misc

Docker/Singularity container

Docker

Available from nigyta/dfast_vrl
Includes NCBI VADR and other dependencies. Modified from staphb/vadr.

Example:
In the example below, docker run is invoked from the application root directory (where the dfast_vrl script exists).

docker run -v $PWD:/data -it --rm nigyta/dfast_vrl:latest dfast_vrl -i /data/examples/LC570964-6.draft.contigs.fa -m /data/examples/metadata.txt -o /data/OUT

Singularity

In the NIG-Super computer system, Singularity container is available at /lustre6/public/vrl/dfast_vrl:latest.sif.
Sample data is available from github.

• assembled contig FASTA

curl -O https://raw.githubusercontent.com/nigyta/dfast_vrl/main/examples/SRR10903401.meta_vrl.contig.fa

• metadata file

curl -O https://raw.githubusercontent.com/nigyta/dfast_vrl/main/examples/metadata.txt

Example:

singularity exec /lustre6/public/vrl/dfast_vrl:latest.sif dfast_vrl -i SRR10903401.meta_vrl.contig.fa -m metadata.txt -o dfv_result

vadr2mss.py

Pipeline for various kind of viruses.

• Internally invokes VADR
• Converts the result of VADR into DDBJ MSS file
• Requires VADR model data to be downloaded before running the pipeline (with download_vadr_models.sh script)

Usage:

vadr2mss.py -i INPUT.fasta -m METADATA.txt -o OUT_DIR -M MODELNAME(e.g. mpox)

To test vadr2mss.py using the Docker/Singularity container, download the sample files from github.

git clone https://github.com/nigyta/dfast_vrl.git
cd dfast_vrl

Download VADR model data

mkdir vadr_models
docker run -it --rm -v $PWD/vadr_models:/vadr_models nigyta/dfast_vrl:latest /dfast_vrl/download_vadr_models.sh

Example using Docker:

docker run -it --rm -v $PWD/vadr_models:/vadr_models -v $PWD/examples:/data nigyta/dfast_vrl:latest vadr2mss.py -i /data/mpox/LC756923.fasta -m /data/mpox/metadata_mpox.txt -o /data/OUT -M mpox

Example using Singularity (NIG supercomputer)

singularity exec -B $PWD/vadr_models:/vadr_models /usr/local/shared_data/vrl/dfast_vrl:latest.sif vadr2mss.py -i examples/mpox/LC756923.fasta -m examples/mpox/metadata_mpox.txt -o OUT_DIR -M mpox