ChIA-PET2报错合集

文章目录

• 一、关于macs2报错：
• • • 1、报错
    • 解决
    • 2、报错
    • 解决
• 二、关于Running Step 6: QCplots报错：
• • • 报错
    • 修改：
• 三、关于第七步报错：
• • • 解决：
• 四、第七步再次报错：

一、关于macs2报错：

1、报错

解决

原因：没在sun conda环境下跑

切到sun环境下就好了

2、报错

在网上找到了一样的报错

Traceback (most recent call last):
  File "/home/server/shm/conda/condaa/envs/sun/bin/macs2", line 653, in <module>
    main()
  File "/home/server/shm/conda/condaa/envs/sun/bin/macs2", line 49, in main
    from MACS2.callpeak_cmd import run
  File "/home/server/shm/conda/condaa/envs/sun/lib/python3.8/site-packages/MACS2/callpeak_cmd.py", line 23, in <module>
    from MACS2.OptValidator import opt_validate
  File "/home/server/shm/conda/condaa/envs/sun/lib/python3.8/site-packages/MACS2/OptValidator.py", line 20, in <module>
    from MACS2.IO.Parser import BEDParser, ELANDResultParser, ELANDMultiParser, \
  File "MACS2/IO/Parser.pyx", line 25, in init MACS2.IO.Parser
  File "MACS2/IO/FixWidthTrack.pyx", line 27, in init MACS2.IO.FixWidthTrack
  File "MACS2/Pileup.pyx", line 19, in init MACS2.Pileup
  File "MACS2/IO/BedGraph.pyx", line 26, in init MACS2.IO.BedGraph
ImportError: /home/server/shm/conda/condaa/envs/sun/lib/python3.8/site-packages/MACS2/IO/ScoreTrack.cpython-38-x86_64-linux-gnu.so: undefined symbol: __log_finite
           

Collecting macs2
  Using cached https://files.pythonhosted.org/packages/21/0f/972b44c84d85e37d816beae88aa5ddad606bd757630d77dc2f558900a6ce/MACS2-2.2.6.tar.gz
    Complete output from command python setup.py egg_info:
    CRITICAL: Python version must >= 3.6!
           

python2.7，3.5，3.7，3.6，3.8都不行

解决

历时两天，终于弄好啦

python3.8下，安装macs2 2.2.5版本的，就没有这个报错了

用这个命令测试macs2，没有报错就ok

二、关于Running Step 6: QCplots报错：

报错

修改：

sudo ln -s /home/server/shm/ruanjian/ChIA-PET2-0.9.3/bin/QCplots.R /usr/bin
           

三、关于第七步报错：

[07-25 16:37:20] Running Step 6: QCplots ...

Rscript /usr/bin/QCplots.R OUTdir6 index-1
载入需要的程辑包：ggplot2
载入需要的程辑包：scales
null device 
          1 

[07-25 16:37:22] Running Step 7: MICC ...

Rscript /usr/bin/MICC2.R OUTdir6/index-1.interactions.intra.bedpe OUTdir6/index-1.interactions.inter.bedpe OUTdir6/index-1.interactions.MICC 2
Fatal error: ??????'/usr/bin/MICC2.R': ?????????

           

搜到的解决办法：

但遇到问题：

Package is not available (for R version XXX)？
           

千万不要随便升级和改动R，或者python，根据这个网址，换了个下载网址，就可以了

https://www.cnblogs.com/jessepeng/p/14301293.html

解决：

sudo ln -s /home/server/shm/ruanjian/ChIA-PET2-0.9.3/bin/MICC2.R /usr/bin
           

四、第七步再次报错：

[07-26 03:06:04] Running Step 7: MICC ...

Rscript /usr/bin/MICC2.R OUTdir7/index-1.interactions.intra.bedpe OUTdir7/index-1.interactions.inter.bedpe OUTdir7/index-1.interactions.MICC 2
载入需要的程辑包：VGAM
载入需要的程辑包：stats4
载入需要的程辑包：splines
Running MICC...
Intra data:  OUTdir7/index-1.interactions.intra.bedpe 
Inter data:  OUTdir7/index-1.interactions.inter.bedpe 
Output file:  OUTdir7/index-1.interactions.MICC 
PET count cutoff:  2 
Minimun Confident PET count:  5 
reltol:  1e-08 
Loading intra data...
Loading inter data...
Cacluating...
Intra:  421 	Inter: 11 
Error in optim(InitialValue, fn = Loglik.theta12.params, gr = D1.Loglik.theta12.params,  : 
  L-BFGS-B不能有'fn'的无限值
Calls: MICCoutput2 ... MICCMainLearn -> EMIter -> Solve.Theta12.params -> optim
此外: Warning messages:
1: In dhyper(cAB, cA, 2 * N - cA, cB, log = TRUE) : 产生了NaNs
2: In dhyper(0, cA, 2 * N - cA, cB) : 产生了NaNs
停止执行

           

在这里单独运行micc，测试到换参数可以跑成功，原因可能是数据太少，解决办法是换大数据。

换参数后可能精确度不够高，待开学测试一百多g的ctcf数据，再看看是否需要换micc参数。