WGAVarHunter (Beta version)

Fast and accurate genetic variation identification through whole genome alignment

Welcome to Whole Genome Alignment-based Variation Hunter (WGAVarHunter）

Basically, the program was written in Rust and it needs samtools, sequence aligner minimap2, winnowmap, unimap or wfmash. Users need to install the required tools before using WGAVHunter.

Pre-compiled version

Here I have attached a pre-compiled edition under Linux and macOS environments.

• In Linux environment, users may need to check glibc by using ldd --version. If the version is 2.17 or above, users can freely use the binary tool.

• In macOS environment, it was compiled in macOS Monterey.

Execution

Users can check help by using WGAVHunter -h

----------------------------------------------------------------------------------------------------
Program: WGAVHunter
Version: 0.1.0
Author:  Andy Yuan ([email protected])
----------------------------------------------------------------------------------------------------
Synopsis: Discover genomic variants based on whole genome alignment through an efficient way

USAGE:
    WGAVHunter [OPTIONS] -r <REFERENCE> -q <QUERY>... -o <OUTDIR>

OPTIONS:
    -r <REFERENCE>                 A reference fasta file
    -q <QUERY>...                  Query fasta file(s). Can be single or multiple
    -n <N_PLOIDY>                  Ploidy level of the species [default: 2]
    -w <WINDOW_SIZE>...            Window size(s) used to split the query fasta (kb). Can be single
                                   or multiple values [default: 500]
    -P <PERCENTAGE>                Percentage (%) of adjacent windows overlapped [default: 10]
    -a <ALIGNER>                   Aligner (minimap2|winnowmap|unimap|wfmash) [default: minimap2]
    -A <ALIGNER_SETTINGS>          Aligner parameter settings in "" [default: "-x asm20"]
    -u <USE_SPLIT>                 Use 'split-prefix' for (minimap2|winnowmap). Could be pretty slow
                                   and storage demanding if the genome size is big [default: false]
    -c <CHUNK_SIZE>                Chunk size (kb) used to parse each chromosome [default: 1000]
    -R <REMOVE_UNQUALIFIED>        Remove query seq with less than n (bp) aligned [default: 1000]
    -m <MAP_QUALITY>               Mapping quality used for variant calling [default: 30]
    -s <CALL_SNVS>                 Call single nucleotide variants (SNVs) [default: true]
    -I <CALL_SMALL_INDELS>         Call small indels [default: true]
    -S <CALL_SVS>                  Call structural variants (SVs) [default: true]
    -N <NOVEL_REGIONS>             Report novel genomic regions in the input fastas [default: true]
    -M <MAX_INDEL_SIZE>            Maximum small indel size (bp) called [default: 49]
    -d <DEFAULT_SV_SIZE>           Minimum SV size (bp) called [default: 50]
    -D <DIST_DUP>                  Maximum allowed distance (bp) between aligned query coordinates
                                   for duplication calling [default: 1000]
    -T <TRANS_SIZE>                Minimum translocation size (kb) called [default: 10]
    -t <THREADS>                   Number of threads [default: 4]
    -p <PREFIX>                    Prefix of the output files [default: WGAVHunter]
    -o <OUTDIR>                    Output directory
    -i <INTER_DIR>                 Intermediate folder [default: $OUTDIR/tmp]
    -k <KEEP_INTER>                Keep intermediate folder and content [default: false]
    -e <ENABLE_DEBUG>              Enable debug mode [default: false]
    -h, --help                     Print help information
    -V, --version                  Print version information

Users may run the program like

./WGAVHunter -r ref.fa -q qry.fa -o .

Note: Parameter settings can be adjusted based on the resources.

Q&A

1. Which aligner should I use?

Depending on the purpose and the resources, users may select one of the aligners recommended in WGAVarHunter to perform the alignment. However, for large genome comparisons (for example genome size > 4Gb), we recommend using unimap or wfmash.

Note: WGAVarHunter relies on the alignment to call variants. As the performance of each aligner differs, users need to decide which one or some should be used.

2. How to set a window size ?

Aligning entire chromosomes directly may take a long time and many computational resouces. To speed up alignment, we enable a function to chop the entire chromosome into smaller pieces. From benchmarking, the results were similar but the running time could reduce significantly.

Note: If users don't want to chop the query genome, set the window size >= the max length of a chromosome in the query fasta. However, due to the limitation of BAM format, the window size (-w) cannot be over 256 Mb for large genomes.

3. What are novel genomic regions ?

Novel genomic regions refer to genomic regions having no sequence aligned. They could be novel sequences or gaps in either reference or query (different from insertions).

Notification

• Variants called from duplication or duplication + inversion regions might be incorrect. Users may need to further filter to meet the research demand.