安装基因组注释流程braker2
github链接
https://github.com/Gaius-Augustus/BRAKER
参考链接
https://www.jianshu.com/p/e6a5e1f85dda
github主页上的安装流程看着还挺麻烦的
参考链接里提到可以用conda安装
我就试着用conda安装
新建一个虚拟环境
conda create -n braker2
conda activate braker2
conda install braker2
这一步卡了一晚上也没有成功,看到有人说可以用mamba试一下
mamba install braker2
这个可以,但是遇到了一个报错
Could not solve for environment specs
Encountered problems while solving:
- package libmambapy-1.3.0-py311h1f88262_0 requires openssl >=3.0.7,<4.0a0, but none of the providers can be installed
我这个虚拟环境下的python是3.11,我换成3.10以后就没有这个报错了
conda install python=3.10 -y
然后再重新安装
mamba install braker2
安装结束以后会提示
The config/ directory from AUGUSTUS can be accessed with the variable AUGUSTUS_CONFIG_PATH.
BRAKER2 requires this directory to be in a writable location, so if that is not the case, copy this directory to a writable location, e.g.:
cp -r /mnt/shared/scratch/myan/apps/mingyan/Biotools/mambaforge/envs/braker2/config/ /absolute_path_to_user_writable_directory/
export AUGUSTUS_CONFIG_PATH=/absolute_path_to_user_writable_directory/config
Due to license and distribution restrictions, GeneMark and ProtHint should be additionally installed for BRAKER2 to fully work.
These packages can be either installed as part of the BRAKER2 environment, or the PATH variable should be configured to point to them.
The GeneMark key should be located in /home/myan/.gm_key and GENEMARK_PATH should include the path to the GeneMark executables.
done
接下来还得手动配置GeneMark
在这个链接 http://exon.gatech.edu/GeneMark/license_download.cgi
![](https://img.haomeiwen.com/i6857799/ecff98fa5bd72c42.png)
这边kernel 2.6 和3.10我暂时搞不清楚有啥区别,我选择的是3.10
![](https://img.haomeiwen.com/i6857799/6887a0fe856e045f.png)
需要填这些信息,然后同意获得下载链接,
关于GeneMark的配置,可以参考 https://github.com/Gaius-Augustus/BRAKER
![](https://img.haomeiwen.com/i6857799/53c85097b2910ae8.png)
尤其是最后一步,不知道为啥如果不运行这一步,不会调用conda环境中的perl,然后就一直报错
在 ~/.bashrc 文件里添加
export GENEMART_PATH=/home/myan/biotools/gmes_linux_64_4
然后进入gmes_linux_64_4这个文件夹
运行
perl change_path_in_perl_scripts.pl "/home/myan/anaconda3/envs/braker2/perl"
这应该就配置好了
然后运行注释命令
braker.pl --genome=at_chr1.fa --bam=SRR4420293.sorted.bam
我这个只是拟南芥一条染色体的数据,如果需要这个数据可以给我留言
这一步需要好长时间,
正常第一步应该先屏蔽重复序列,这里我忘记做这一步了
参考这个
https://xuzhougeng.top/archives/genome-annotation-with-braker2
关于conda的一个新的知识点
不是一定要用conda activate 启动环境,才能调用命令,你其实可以调用某个环境的给定指令
conda run -n rna-seq STAR --help
参考 https://xuzhougeng.top/archives/activate-conda-environment-in-shell-script