EPIC stops if treatment and control are both input files

[c-guzman]

Because the way my pipeline is set up, it will typically run Input bed vs Input bed ... in MACS2 this is fine ... no peaks are called. But EPIC gets a command exit status 1:

Here is the output:

Command executed:

  epic --treatment Input_treatment.bed --control Input_control.bed --number-cores 3 --genome hg19 --fragment-size 150 > Input_epic.bed

Command exit status:
  1

Command output:
  (empty)

Command error:
      710 
      711     rizer = klass(max(len(lk), len(rk)))
      712 
  --> 713     llab = rizer.factorize(lk)
      714     rlab = rizer.factorize(rk)
      715 
      716     count = rizer.get_count()
      717 

  ...........................................................................
  /usr/local/anaconda2/lib/python2.7/site-packages/pandas/hashtable.so in pandas.hashtable.Int64Factorizer.factorize (pandas/hashtable.c:15715)()
      854 
      855 
      856 
      857 
      858 
  --> 859 
      860 
      861 
      862 
      863 

  ...........................................................................
  /usr/local/anaconda2/lib/python2.7/site-packages/pandas/hashtable.so in View.MemoryView.memoryview_cwrapper (pandas/hashtable.c:29784)()
      639 
      640 
      641 
      642 
      643 
  --> 644 
      645 
      646 
      647 
      648 

  ...........................................................................
  /usr/local/anaconda2/lib/python2.7/site-packages/pandas/hashtable.so in View.MemoryView.memoryview.__cinit__ (pandas/hashtable.c:26059)()
      340 
      341 
      342 
      343 
      344 
  --> 345 
      346 
      347 
      348 
      349 

  ValueError: buffer source array is read-only
  ___________________________________________________________________________

[endrebak]

This is because of a bug in pandas. Try it without multiple cores :)

On Thu, Aug 11, 2016 at 8:57 PM, Carlos Guzman notifications@github.com wrote:

Because the way my pipeline is set up, it will typically run Input bed vs Input bed ... in MACS2 this is fine ... no peaks are called. But EPIC gets a command exit status 1:

Here is the output:

Command executed:

epic --treatment Input_treatment.bed --control Input_control.bed --number-cores 3 --genome hg19 --fragment-size 150 > Input_epic.bed

Command exit status: 1

Command output: (empty)

Command error: 710 711 rizer = klass(max(len(lk), len(rk))) 712 --> 713 llab = rizer.factorize(lk) 714 rlab = rizer.factorize(rk) 715 716 count = rizer.get_count() 717

........................................................................... /usr/local/anaconda2/lib/python2.7/site-packages/pandas/hashtable.so in pandas.hashtable.Int64Factorizer.factorize (pandas/hashtable.c:15715)() 854 855 856 857 858 --> 859 860 861 862 863

........................................................................... /usr/local/anaconda2/lib/python2.7/site-packages/pandas/hashtable.so in View.MemoryView.memoryview_cwrapper (pandas/hashtable.c:29784)() 639 640 641 642 643 --> 644 645 646 647 648

........................................................................... /usr/local/anaconda2/lib/python2.7/site-packages/pandas/hashtable.so in View.MemoryView.memoryview.cinit (pandas/hashtable.c:26059)() 340 341 342 343 344 --> 345 346 347 348 349

ValueError: buffer source array is read-only

---

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/endrebak/epic/issues/36, or mute the thread https://github.com/notifications/unsubscribe-auth/AQ9I0hZamm-zmcuSCcPjWPlRbdyvT0Wuks5qe3CwgaJpZM4Jib94 .

[endrebak]

Did you get more or a longer error message than the one you posted? Did it work with one core?

[endrebak]

Version 0.1.12 should fix this. Thanks for reporting.

pip install bioepic==0.1.12

[c-guzman]

Just wanted to update this and say that the no cores option did fix the issues. Thanks!

[endrebak]

Okay, super. If you also tried version 0.1.12 and it did not work, please tell me :)