rad_assembler

Nextflow pipeline for reference assisted de novo assembling whole genome shotgun sequenced viruses.

Usage

Install Nextflow (>=21.10.3)

Install Docker

Command line:

nextflow run greninger-lab/rad_assembler \
    --input PATH_TO_SAMPLE_CSV \                      # required
    --outdir PATH_TO_OUTPUT_FOLDER \                  # required
    --bowtie2_host_index PATH_TO_HOST_BOWTIE2_INDEX \ # optional (path to bowtie2 index of host to use for filtering host DNA)
    --region_map PATH_TO_REGION_MAP_FILE \            # optional (path to region map json file)
    --genbank_ref ../NC_XXXXXXXX.gb \                 # required (genbank reference)
    --spades_flag \                                   # optional (default is meta, alternatives are careful or unicycler)
    -profile docker \                                 # required
    -with-tower \                                     # optional (use if you want to use Nextflow Tower)
    -c nextflow_aws.config \                          # optional (AWS account config info) 
    -r main                                           # required (use the github main branch)

Region map files:

An optional region map in json format for splitting the reference into sections so that a new reference for read mapping can be built more accurately. This can be very helpful when there are large inverted repeat regions in the genome.

Example region map file

{
    "regions": {
        "TRL" : [0,9212],
        "UL" : [9213,117159],
        "IRL_IRS" : [117160,132604],
        "US" : [132605,145588],
        "TRS" : [145589,152221]
    },
    "region_order": [
        {"region": "TRL", "reverse": false}, 
        {"region": "UL", "reverse": false}, 
        {"region": "IRL_IRS", "reverse": false}, 
        {"region": "US", "reverse": false},
        {"region": "TRS", "reverse": false}
    ]
}

"regions" defines a region name and the reference coordinates for extracting the region. De novo assembled scaffolds are mapped to the region and then a new reference is generated for the region. "region_order" sets the order for concatenating the new region references together to make a complete genome reference. "reverse" will reverse compliment the region reference before concatenating if set to true.

Example region map files:

rad_assembler/region_maps/HSV1-NC001806.json
rad_assembler/region_maps/HSV2-NC001798.json

Sample csv example:

rad_assembler/assets/example.csv

Sample csv format:

sample,fastq_1,fastq_2
F79217_S26,sample_L001_R1_001.fastq.gz,sample_L001_R2_001.fastq.gz

You can create a sample csv file from an s3 folder (including single depth subdirectories) using the shell script generateawssample_csv.sh in bin folder, like this:

generate_aws_sample_csv.sh s3://bucket-name /folder/path/to/run/folder/ csv_name _L001_R1_001 _L001_R2_001

note: replace L001R1001 _L001R2_001 with the suffixes of read1 and read2 if necessary