VPrimer

A method of designing and updating primer and probe with high variant coverage for RNA virus detection

1. Get VPrimer

To get VPrimer, follow the queries.

$ git clone https://github.com/qhtjrmin/VPrimer.git

The source codes are in "src" folder. Since the VPrimer consists of several steps, the vPrimer must be executed using a different input for each step. The guidline for execution is in section 2.

Environment

VPrimer needs the following softwares to run in the system:

• CUDA toolkit version 8 or higher.
• Nvidia driver (v384 or higher)
• gcc/g++ 4.8.x or later

2. Prepare data

2.1. Prepare input sequence data

The example of input files that can be used are in "input_test" folder. In the case of host data, it should be decompressed as the following command to use for input.

$ gunzip human_host.txt.gz

The required data is as follow:

• host sequence
• virus sequence
• mapping data for virus sequence (oid-vidset, sid-oid, sid-vidset)

The mapping data is for coverage checking and homology test against other virus. Refer to test inputs and the paper for the format and meaning of the files.

2.2. Input data generation

Refer to the following link: https://github.com/Hajin-Jeon/VPrimer-input-generator

2.3. Get VPrimer execution file

The basic execution order is as follows.

• find host subsequences (host_primer and host_probe)
• find virus primer and probe candidates (virus_primer and virus_probe)
• variant coverage filtering: filter out sequences under 95% coverage (check_coverage)
• find primer pair having at least one proper probe (primer_pair) (not in GitHub)
• find final primer-probe sets (final_set) (not in GitHub)

You can run additional check_coverage, referring to the paper.

3. Run

3.1. Generate executable files

You can select one of these two methods: (1) run build.sh in src folder (2) run this command in each folders in src folder.

$ nvcc -O3 -Xcompiler -fPIC --compile --relocatable-device-code=false -gencode arch=compute_52,code=compute_52 -gencode arch=compute_53,code=compute_53 -gencode arch=compute_60,code=compute_60 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_53,code=sm_53 -gencode arch=compute_60,code=sm_60  -x cu -o  "VPrimer.o" "VPrimer.cu" -std=c++14

$ nvcc --cudart static --relocatable-device-code=false -gencode arch=compute_52,code=compute_52 -gencode arch=compute_53,code=compute_53 -gencode arch=compute_60,code=compute_60 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_53,code=sm_53 -gencode arch=compute_60,code=sm_60 -link -o  "vprimer" ./VPrimer.o -std=c++14

3.2. Run for each function

Example: in virus_primer folder

./vprimer -i ../../input_test/host_with_virus.txt -o output/ -d directory/ -s 5000 -t 35 -g 4

3.3. Usage

run.sh script in src folder can be used to analyze host-virus interactions. The script requires several input files and options to be specified.

3.3.1. Options

• -H <path> : path to the host sequence database file (required)
• -V <path> : path to the virus sequence database file (required)
• -o <path> : path to the organism-vid mapping database file (required)
• -v <path> : path to the sid-vid mapping database file (required)
• -t <num> : number of CPU threads (optional, default: 20)
• -g <num> : number of GPU threads (optional, default: 1)

3.3.2. Example

To run the script with the following input files and options:

• host sequence database file: ../input_test/human_host.txt
• virus sequence database file: ../input_test/test_virus.txt
• organism-vid mapping database file: ../input_test/oid_vidset.txt
• sid-vid mapping database file: ../input_test/sid_vidset.txt
• number of CPU threads: 35
• number of GPU threads: 4

Use the following command:

./run.sh -H ../input_test/human_host.txt -V ../input_test/test_virus.txt -o ../input_test/oid_vidset.txt -v ../input_test/sid_vidset.txt -t 35 -g 4

Then, output data will be stored in check_coverage/output/output

Note: The paths to the input files may vary depending on the actual file locations.