outrigger --index error: "AttributeError: 'Series' object has no attribute 'iteritems'"

[jordimaggi]

Hi,

I am running outrigger for the first time. I am running it on an RNA-seq dataset aligned with STAR, which resulted in a .bam file of 3.6 GB and a corresponding SJ.out.tab file of 14.6 MB.

I used the following command:

outrigger index --sj-out-tab *SJ.out.tab --gtf hg19.ensGene.gtf

I obviously forgot to check if there was an option to use multiple cores to run the command. Therefore, it successfully ran for 5 days, up to the Making ".bed" files for exons in each event ... step, which failed due to the following error:

AttributeError: 'Series' object has no attribute 'iteritems'

Any idea on how to fix this? Also, can I re-start the command without repreating the steps that it was able to complete? If so, how?

Thanks for your help.

The conda environment that I am using to run the command has the following packages installed:

[SOIL-TiME]

Hi, All.
i got same issue.
i tested latest and 1.0.0 version.

[shdam]

Hey,

In case people are still running into this issue - it appears to be because Outrigger requires an older version of Python. The issue was solved for me with this environment.yml:

name: outrigger-env
channels:
  - conda-forge
  - r
  - bioconda
dependencies:
- python=3.6.7
- pytest
- pandas>=0.17.0
- coverage
- gffutils
- pybedtools
- biopython
- bedtools
- joblib
- pysam
- sphinx>=1.3.6
- sphinx_rtd_theme
- outrigger
- pip:
    - graphlite
    - recommonmark==0.4.0
    - nbsphinx

I built the environment using mamba
mamba env create --file environment.yml

Hope this is helpful.