single-cell-genetics / cellSNP

Pileup biallelic SNPs from single-cell and bulk RNA-seq data
Apache License 2.0
71 stars 11 forks source link

Problem with New Version (0.1.3) #2

Closed elderni closed 5 years ago

elderni commented 5 years ago

I updated cellSNP yesterday and found that data that had previously worked with the tool no longer did. I also tested the data used in your test.sh file and got the same traceback. Here is an example of the output:

[cellSNP] mode 2: pileup 22 whole chromosomes in 4246 single cells. Traceback (most recent call last): File "/home/rstudio/.local/bin/cellSNP", line 11, in sys.exit(main()) File "/home/rstudio/.local/lib/python2.7/site-packages/cellSNP/cellSNP.py", line 198, in main result = [res.get() if nproc > 1 else res for res in result] File "/usr/lib/python2.7/multiprocessing/pool.py", line 572, in get raise self._value RuntimeWarning: divide by zero encountered in log

I ultimately uninstalled cellSNP and reinstalled version 0.0.8 and found that it works just fine again. I'm still not sure what caused the issue in the newer version but wanted to alert you to it.

huangyh09 commented 5 years ago

Hi, it looks that you are using python 2.7. Could you try it with Python 3. I haven't tested the new versions under Python 2.

Also, cellSNP (>=0.1.0) requires pysam>=0.15.2. Not sure which causes the issue, but your feedback will appreciated.

elderni commented 5 years ago

That solved the problem! Thank you.