installation of 1.4 about cite-seq-count HOT 11 CLOSED

Hello @joonlee3
yes, you can try to use this option: --no-cache-dir
it should clean the local version you have downloaded before.

from cite-seq-count.

Thank you for the quick reply. I followed your suggestion but v1.3.2 still seems to be installed.

{0}: ./pip install CITE-seq-Count --no-cache-dir
Collecting CITE-seq-Count
Downloading https://files.pythonhosted.org/packages/40/f6/a9837f7ce9ef253f496e807636530b961d78dee317685702f79bc2a1a5ca/CITE-seq-Count-1.3.2.tar.gz
Requirement already satisfied: regex>=2018.07.11 in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from CITE-seq-Count) (2019.1.24)
Requirement already satisfied: python-levenshtein>=0.12.0 in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from CITE-seq-Count) (0.12.0)
Requirement already satisfied: pandas>=0.23.3 in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from CITE-seq-Count) (0.24.0)
Requirement already satisfied: setuptools in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from python-levenshtein>=0.12.0->CITE-seq-Count) (40.7.0)
Requirement already satisfied: numpy>=1.12.0 in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from pandas>=0.23.3->CITE-seq-Count) (1.16.0)
Requirement already satisfied: pytz>=2011k in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from pandas>=0.23.3->CITE-seq-Count) (2018.5)
Requirement already satisfied: python-dateutil>=2.5.0 in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from pandas>=0.23.3->CITE-seq-Count) (2.7.3)
Requirement already satisfied: six>=1.5 in /site/ne/home/wings/.pyenv/versions/3.5.2/lib/python3.5/site-packages (from python-dateutil>=2.5.0->pandas>=0.23.3->CITE-seq-Count) (1.11.0)
Installing collected packages: CITE-seq-Count
Running setup.py install for CITE-seq-Count ... done
Successfully installed CITE-seq-Count-1.3.2

from cite-seq-count.

What is your pip version?

from cite-seq-count.

pip 19.0.1 (python 3.5)

from cite-seq-count.

Maybe it's because of python3.5. I would advise 3.6 at least also because of performance and memory issues. Could you try with 3.6?

from cite-seq-count.

I just double checked. Before 1.4.1 it was python3 required and now it's 3.6. That's probably why you get the old version

from cite-seq-count.

Sure, I will try.

Would you please describe the major bug in the previous version in more details? I am using 10x genomics technology. Do you expect some differences if I re-analyze our data by using v1.4?

Thank you so much.
Joon

from cite-seq-count.

Sure.

The bug in 1.3.4 was that the umi grouping was off and this gave an output that was in between reads and umis.

One major difference in the latest version is the umi correction. Small sequencing mistakes might have inflated some umi counts in the past. Depending on your sequencing quality, you might see a "clearer" picture to discriminate cells holding a tag or not.

from cite-seq-count.

Hello, I would like to know if the bug was only in version 1.3.4 or also in previous versions.
Thank you very much
Alice

from cite-seq-count.

Hello,
it was only in 1.3.4. The versions before had non corrected UMI counts.
Can I close this?

from cite-seq-count.

OK, thank you very much.

from cite-seq-count.