CNVkit 安装和使用（一）

1. Python2和Python3共存安装

Linux系统自带python版本为 2.7.5，CNVkit推荐使用python3，因此，在Linux上安装Python 3.9.2 版本

# Linux联网测试
ping www.baidu.com

PING www.wshifen.com (103.235.46.39) 56(84) bytes of data.
64 bytes from 103.235.46.39 (103.235.46.39): icmp_seq=1 ttl=55 time=249 ms
64 bytes from 103.235.46.39 (103.235.46.39): icmp_seq=2 ttl=55 time=249 ms
64 bytes from 103.235.46.39 (103.235.46.39): icmp_seq=3 ttl=55 time=249 ms
64 bytes from 103.235.46.39 (103.235.46.39): icmp_seq=4 ttl=55 time=249 ms

# 下载和安装Python
wget -c https://www.python.org/ftp/python/3.9.4/Python-3.9.4.tar.xz

tar -Jxvf Python-3.9.4.tar.xz

cd Python-3.9.4

# 配置python3安装路径;python2路径为/usr/bin/python2
./configure prefix=/usr/local/python3

# 编译安装python
make && make install

2. CNVKit安装

# CNVkit安装方法1
wget https://github.com/etal/cnvkit/archive/v0.9.7.zip
unzip v0.9.7.zip
cd cnvkit-0.9.7/

# CNVkit安装方法2
git clone https://github.com/etal/cnvkit
cd cnvkit/
/usr/local/python3/bin/pip3 install -e .

# 测试安装是否成功
/usr/local/python3/bin/python3 /opt/software/cnvkit-0.9.7/cnvkit.py -h
/usr/local/python3/bin/python3 /opt/software/cnvkit/cnvkit.py -h

# 报错

缺少pandas.png

3. 安装pandas库

/usr/local/python3/bin/pip3 install pandas

# 报错：pip is configured with locations that require TLS/SSL, however the ssl module in Python is not available.

缺少ssl.png

3. 1 确认缺少ssl模块错误：

在python shell import ssl同样提示找不到ssl模块

证实缺少ssl.png

3.2 解决缺少ssl模块问题：

使用rpm -aq|grep openssl命令查询openssl是否正确安装。发现缺少openssl-devel包

缺少level包.png

# 安装openssl-devel包
yum install -y openssl-devel

openssl-devel 包安装成功.png

3.3 测试问题是否解决

# 重新编译安装python
cd /opt/software/Python-3.9.2
make clean 

# 配置安装路径
./configure prefix=/usr/local/python3

# 编译安装python
make && make install

# 重新安装pandas
/usr/local/python3/bin/pip3 install pandas

# pandas安装成功
/usr/local/python3/bin/pip3 list

pandas安装成功.png

#在python shell中 import ssl同样找到ssl模块

ssl 问题解决.png

4. 安装Bio库

# 成功安装pandas之后，测试CNVkit是否能正常运行，
/usr/local/python3/bin/python3 /opt/software/cnvkit-0.9.7/cnvkit.py -h

# 报错：ModuleNotFoundError: No module named 'Bio'

Bio包报错.png

4.1 解决缺少Bio库问题（此类报错解决原则缺什么补什么）：

# python安装Bio包
/usr/local/python3/bin/pip3 install Bio

# 仍然报错：ModuleNotFoundError: No module named 'Bio'

4.2 继续解决缺少Bio包问题：发现Bio库的安装并不是pip install bio，而是biopython

# 安装biopython
/usr/local/python3/bin/pip3 install biopython

# 还是报错：ModuleNotFoundError: No module named '_ctypes'

_ctypes 报错_version.png

# 搜索分析原因需要安装libffi-devel（参考：https://www.jianshu.com/p/b1f575c45e10）
yum install libffi-devel -y

# 重新编译安装python
cd /opt/software/Python-3.9.2
make clean 

# 配置安装路径
./configure prefix=/usr/local/python3

# 编译安装python
make && make install

# 重新安装 biopython
/usr/local/python3/bin/pip3 install biopython

# biopython安装成功

biopython 安装成功.png

5. 安装pysam、scipy、ReportLab、pyfaidx和pomegranate库

# 成功安装biopython之后，测试CNVkit是否能正常运行，
/usr/local/python3/bin/python3 /opt/software/cnvkit-0.9.7/cnvkit.py -h

# 相同报错持续出现，不断安装即可：ModuleNotFoundError: No module named 'pysam' ；ModuleNotFoundError: No module named 'scipy'等等，不再复制了。

pysam缺少.png 缺少scipy.png 缺少ReportLab.png 缺少pyfaidx.png 缺少pomegranate.png

# 解决办法相同

#[图片上传中...(pysam安装成功.png-48740c-1616666692869-0)]

/usr/local/python3/bin/pip3 install pysam

/usr/local/python3/bin/pip3 install scipy

/usr/local/python3/bin/pip3 install ReportLab

/usr/local/python3/bin/pip3 install pyfaidx

/usr/local/python3/bin/pip3 install pomegranate

pysam安装成功.png 成功安装scipy.png ReportLab安装成功.png 成功安装pyfaidx.png pomegranate安装成功.png

6. 确认CNVkit安装成功

CNVkit 安装成功.png

7. 安装需要的 R依赖包，不安装的话CNVkit运行无法生成 .cns .pdf .png文件

# 安装 R软件
yum install R

# 加载需要的R包
> install.packages("BiocManager")
> library(BiocManager)
> BiocManager::install("DNAcopy")
> q()