SVision is a deep learning-based structural variants caller that takes aligned reads or contigs as input. Especially, SVision implements a targeted multi-objects recognition framework, detecting and characterizing both simple and complex structural variants from three-channel similarity images.
SVision is free for non-commercial use by academic, government, and non-profit/not-for-profit institutions. A commercial version of the software is available and licensed through Xi’an Jiaotong University. For more information, please contact with Jiadong Lin ([email protected]) or Kai Ye ([email protected]).
## Get the source code
git clone https://github.com/xjtu-omics/SVision.git
cd SVision
## Create a conda environment for SVision
conda env create -f ./environment.yml
## Install from source
conda activate svisionenv
python setup.py install
The Pip and Conda install would be available later.
docker pull jiadongxjtu/svision:1.3.6
docker run jiadongxjtu/svision:1.3.6 SVision -h
Note: Please ensure you have the permission to write into docker.
Please visit our wiki page for performance evaluation.
SVision [parameters] -o <output path> -b <input bam path> -g <reference> -m <model path>
SVision -h
Required Input/Ouput parameters
-o OUT_PATH Absolute path to output
-b BAM_PATH Absolute path to bam file
-m MODEL_PATH Absolute path to CNN predict model
-g GENOME Absolute path to your reference genome (.fai required in the directory)
-n SAMPLE Name of the BAM sample name
-g
path to the reference genome, the index file should under the same directory.
-m
path to the pre-trained deep learning model svision-cnn-model.ckpt (external download link).
Please check the wiki page for more usage and parameter details.
The demo data is ./supports/HG00733.svision.demo.bam. The HiFi whole genome sequencing data of HG00733 is published on Science.
-
Download the reference genome GRCh38
-
Run SVision with your reference
SVision -o ./output_dir -b ./supports/HG00733.svision.demo.bam -m /path/to/svision-cnn-model.ckpt -g ./reference.fa -n HG00733 -s 5 --graph --qname
Please use the same parameter settings if you use the docker image.
- Output files
*.graph.vcf
The standard VCF output with CSV graph info columns.
*.graph_exactly_match.txt
CSV graphs of exactly identical structure.
*.graph_symmetry_match.txt
Identified isomorphic graphs from all CSV graphs.
graphs
The directory for CSV graph in rGFA format.
Please check the wiki page for more details of output format.
If you have any questions, please feel free to contact: [email protected], [email protected]