由于对cfDNA的兴趣,查到22年这篇在Nature biotechnology上的文章Inferring gene expression from cell-free DNA fragmentation profiles.想尝试跑一下EPIC-seq的流程。感谢文章作者提供的这么全面的资料。但,仍然很多不懂。
无论用conda安装R,还是下载gz文件本地安装,都是失败。
太难过了。
一、 Rscript 这个是在Linux里面使用R要调用的命令。
Rscript runEPIC.R –-bamdir D:/work/research/ssDNA/result/ssDNA_1stsequence/bamdir –-tssinfo D:/work/research/ssDNA/result/ssDNA_1stsequence/code/priordata/all.tss.genes.canonical.ensembl75.txt –-panelbed D:/work/research/ssDNA/result/ssDNA_1stsequence/code/priordata/all.tss.genes.canonical.ensembl75.txt.noheader –-targeted no –-outdir D:/work/research/ssDNA/result/ssDNA_1stsequence/outdir –-mapq 30 –-groupref <character> –-groupsize <double, number of genes per group> –-skipHist <binary, 0 or 1> –-skipOCFFrags <binary, 0 or 1>
二、先建立虚拟环境
conda create -n R_cfDNApipe python=3.8.3
输入:conda activate R_cfDNApipe,报错。
于是先输入:source activate,再输入 conda activate R_cfDNApipe
查看所有虚拟环境 conda env list
image.png
退出虚拟环境: conda deactivate
参考教程: 虚拟环境
虚拟环境:https://blog.csdn.net/robot8me/article/details/109471568
conda的配置:https://www.bilibili.com/read/cv8956636
conda瘦身:
conda clean --packages --tarballs
删除缓存包,重新下载
conda clean -i
清除缓存,重新加载condo地址
conda clean -p //删除没有用的包
conda clean -t //删除tar包
conda clean -y --all //删除所有的安装包及cache
三、在Linux里直接安装R(直接安装失败,就是到PCRE那一步怎么都不成功)
参考教程:
https://www.136.la/jingpin/show-30899.html
https://blog.csdn.net/tandelin/article/details/1183937531
https://www.cnblogs.com/jessepeng/p/10984983.html
下载文件 R-4.1.1.tar.gz,放到software目录下
2,解压:tar -zxvf R-4.1.1.tar.gz
3,进入解压后的文件夹
进行configure,make, make install
./configure --prefix=/data/zds209/software/R --enable-R-shlib
make
make install
4,configure遇到各种报错:
-
zlib version >= 1.2.5需要安装
由于公共software里有,只需直接.bashrc添加:
export PATH=/data/software/zlib1.22.11/bin:$PATH
export CFLAGS="-fPIC -I/data/software/zlib1.22.11/include"
export LDFLAGS="-fPIC -L/data/software/zlib1.22.11/lib" -
bzip2需要安装
下载https://sourceforge.net/projects/bzip2/files/latest/download
tar xzvf bzip2-1.0.6.tar.gz
cd bzip2-1.0.6
make -f Makefile-libbz2_so
make clean
make
make install PREFIX=/data/zds209/software/bzip2-1.0.6
- lzma version >= 5.0.3
下载http://tukaani.org/xz/xz-5.2.3.tar.gz
tar xzvf xz-5.2.3.tar.gz./configure --prefix=5.
cd tar xzvf xz-5.2.3.tar.gz
./configure --prefix=/data/zds209/software/xz-5.2.3
make -j3
make install
tar -zxvf pcre-8.42.tar.gz
cd pcre-8.42
./configure --enable-utf8 --prefix=/data/zds209/software/pcre-8.42
make -j3
make install
到.bashrc中
export LC_ALL="zh_CN.UTF-8"
export LANG="zh_CN.UTF-8"
export PATH=/data/zds209/software/pcre-8.42/bin:$PATH
export LD_LIBRARY_PATH=/data/software/pcre-8.42/bin:$LD_LIBRARY_PATH
查看一下pcre的版本:
zds209 07:14:44 ~locale
LANG=zh_CN.UTF-8
LC_CTYPE="zh_CN.UTF-8"
LC_NUMERIC="zh_CN.UTF-8"
LC_TIME="zh_CN.UTF-8"
LC_COLLATE="zh_CN.UTF-8"
LC_MONETARY="zh_CN.UTF-8"
LC_MESSAGES="zh_CN.UTF-8"
LC_PAPER="zh_CN.UTF-8"
LC_NAME="zh_CN.UTF-8"
LC_ADDRESS="zh_CN.UTF-8"
LC_TELEPHONE="zh_CN.UTF-8"
LC_MEASUREMENT="zh_CN.UTF-8"
LC_IDENTIFICATION="zh_CN.UTF-8"
LC_ALL=zh_CN.UTF-8
用which查位置,发现不对劲,怎么conda里面也有。赶紧到.bashrc里面把conda的PATH注释掉。也不行。
于是进入到conda虚拟环境中,conda安装PCRE。
到R-4.1.1目录下。
./configure --prefix=/data/zds209/software/R --enable-R-shlib --with-readline=yes --with-libpng=yes --with-blas --with-pcre1 LDFLAGS="-L /data/zds209/software/bzip2-1.0.6/lib -L /data/zds209/software/xz-5.2.3/lib -L /data/zds209/software/bzip2-1.0.6/include -L /data/zds209/software/xz-5.2.3/include -L /data/zds209/software/pcre/lib -L /data/zds209/software/pcre/include"
四、虚拟环境中安装R
https://www.jianshu.com/p/b9eb874fc8f4
conda clean -i #清缓存
conda create -n R_EPIC-seq python=3.8.3 #新建一个虚拟环境
source activate
conda activate R_EPIC-seq
conda search R
conda install r-base=4.0#总是在solving environment
仍然失败。。。。。
教程:
https://yunwei365.blog.csdn.net/article/details/108126550?spm=1001.2101.3001.6650.12&utm_medium=distribute.pc_relevant.none-task-blog-2~default~OPENSEARCH~default-12.no_search_link&depth_1-utm_source=distribute.pc_relevant.none-task-blog-2~default~OPENSEARCH~default-12.no_search_link&utm_relevant_index=18
网友评论