从宏基因/转录组数据中识别病毒信号缺少1. 广谱基因marker 2. 数据库代表 3. 有效工具。但VirSorter2很OK。
标题：VirSorter2: a multi-classifier, expert-guided approach to detect diverse DNA and RNA viruses
中文：一种含多分类器且有专家指导的DNA/RNA病毒检测方法
杂志：Microbiome
时间：2021
VirSorter2 Github: https://github.com/jiarong/VirSorter2
bitbucker: https://bitbucket.org/MAVERICLab/virsorter2/src/master/
安装
conda create -n virsort
conda activate virsort
conda install virsorter=2
virsorter --version
#[2021-09-16 09:14 INFO] VirSorter 2.2.3
virsorter --help
Usage: virsorter [OPTIONS] COMMAND [ARGS]...
virsorter - workflow for identifying viral sequences
Options:
--version Show the version and exit.
-h, --help Show this message and exit.
Commands:
config subcommand for configuration management
run run virsorter main workflow
setup download reference files (~10GB) and install dependencies
train-feature subcommand for training feature of customized classifier
train-model subcommand for training customized classifier model
数据库
数据链接：https://osf.io/v46sc/download
win下载
tar -zxvf db.tgz
cd db
数据库中有那些信息呢：
- conda_envs：空的
- group：marker gene和模型
-
hmm：各物种蛋白家族数据库（存储最大）
-
rbs：不知道是啥？
配置数据库：
virsorter config --init-source --db-dir=/route/database/virsort/db/
# /hwfssz5/ST_META/P18Z10200N0423_ZYQ/MiceGutProject/hutongyuan/software/miniconda3/envs/virsort/lib/python3.6/site-packages/virsorter/template-config.yaml
[2022-12-19 20:41 INFO] VirSorter 2.2.3
[2022-12-19 20:41 INFO] /public/home/zzumgg03/huty/softwares/miniconda3/envs/virsort/bin/virsorter config --init-source --db-dir=/public/home/zzumgg03/huty/databases/vs2/db/
[2022-12-19 20:41 INFO] saving /public/home/zzumgg03/huty/databases/vs2/db as DBDIR to config file /public/home/zzumgg03/huty/softwares/miniconda3/envs/virsort/lib/python3.10/site-packages/virsorter/template-config.yaml
测试
virsorter run \
--min-length 1500 \
-j 4 all \
--verbose \
-i ./AF01-22.fna \
-w ./test.out
## demo
virsorter run -w out -i all.fa --min-length 1000 -j 16 all
参数：
-w：输出文件
-i：输入文件
--min-length：最短contig，默认0
-j：并行job数，默认48
all：3 steps: 1) preprocess, 2) feature extraction, and 3) classify
classify：only run 3) classify
--keep-original-seq:
keep the original sequences instead of trimmed; By default, the untranslated regions at both ends of identified viral seqs are trimmed; circular sequences are modified to remove overlap between both ends and adjusted for the gene splitted into two ends;
--include-groups:
dsDNAphage,NCLDV,RNA,ssDNA,lavidaviridae [default: dsDNAphage,ssDNA]
--use-conda-off：Stop using the conda envs (vs2.yaml)
第一次使用需要联网让conda安装依赖
第二次使用可以不连网
运行需要conda，故--use-conda-off参数不能用
结果
病毒类型，序列位置，序列，完整度，打分等等
final-viral-score.tsv
||{i}_partial: 除此provirus序列还有其他provirus/host序列
||full: 可能是provirus或者provirus的一部分，也可能是free virus
||lt2gene: less than 2 genes 在contig中
sequence name
score of each viral sequences across groups (multiple columns)
max score across groups
max score group
contig length
hallmark gene count
viral gene %
nonviral gene %
final-viral-boundary.tsv
seqname：序列名称
trim_orf_index_start
trim_orf_index_end
trim_bp_start：切割起点
trim_bp_end：切割终点
trim_pr
trim_pr_max
prox_orf_index_start
prox_orf_index_end
prox_bp_start
prox_bp_end
prox_pr
prox_pr_max
partial
full_orf_index_start
full_orf_index_end
full_bp_start
full_bp_end
pr_full arc
bac
euk
vir
mix
unaligned
hallmark_cnt
group：病毒分类 ssDNA dsDNAphage
shape：病毒基因组形状 linear circular
seqname_new：新序列名称
病毒鉴定SOP
QC SOP: https://www.protocols.io/view/viral-sequence-identification-sop-with-virsorter2-btv8nn9w
更多：
output detail
