MAJIQ deltapsi index error

[Wanru Lin]

Hello,

I met this error when I tried to run majiq deltapsi:

majiq deltapsi \

-psi1 MAJIQ_PsiCoverage/1A MAJIQ_PsiCoverage/1B MAJIQ_PsiCoverage/1C \

-psi2 MAJIQ_PsiCoverage/6A MAJIQ_PsiCoverage/6B MAJIQ_PsiCoverage/6C \

--names exp1 exp2 \

--min-experiments 3 --splicegraph MAJIQ_build/splicegraph.zarr

----------------------------

2025-07-25 14:37:16,021 (ERROR) - Exiting due to exception:
Traceback (most recent call last):

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/rna_majiq/_run/_run.py", line 147, in run

    self._run(args)

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/rna_majiq/_run/deltapsi.py", line 231, in run

    df = deltapsi_voila.to_dataframe(

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/rna_majiq/voila/DeltaPsiDataset.py", line 365, in to_dataframe

    "raw_psi_mean": self.groups.raw_psi_mean,

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/rna_majiq/core/MixinPsiInference.py", line 76, in raw_psi_mean

    return self.raw_posterior_mean

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/functools.py", line 981, in __get__

    val = self.func(instance)

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/rna_majiq/core/MixinPsiInference.py", line 60, in raw_posterior_mean

    return self.raw_alpha / (self.raw_alpha_plus_beta)

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/xarray/core/_typed_ops.py", line 258, in __truediv__

    return self._binary_op(other, operator.truediv)

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/xarray/core/dataarray.py", line 4721, in _binary_op

    self, other = align(self, other, join=align_type, copy=False)

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/xarray/core/alignment.py", line 882, in align

    aligner.align()

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/xarray/core/alignment.py", line 573, in align

    self.assert_no_index_conflict()

  File "/home/userX/anaconda3/envs/rna_seq/lib/python3.10/site-packages/xarray/core/alignment.py", line 318, in assert_no_index_conflict

    raise ValueError(

ValueError: cannot re-index or align objects with conflicting indexes found for the following coordinates: 'grp' (2 conflicting indexes)

Conflicting indexes may occur when

- they relate to different sets of coordinate and/or dimension names

- they don't have the same type

- they may be used to reindex data along common dimensions

------------------------------

I also got the same error when I was trying to run the workshop sample data from the MAJIQ website. 

I'd appreciate any suggestion about what is causing this and how to fix it!

Thank you very much in advance,

Wanru

[San Jewell]

Hi Wanru,

Using the workshop example data, which is the series of commands you are using in the processing pipeline up to this point?

Thanks,

-San

[Wanru Lin]

Hi San,

I first create a experiments_info.tsv file, which includes information: 

path group

./workshop_example/bam/workshop_Adr1.bam Adr

./workshop_example/bam/workshop_Adr2.bam Adr

./workshop_example/bam/workshop_Adr3.bam Adr

./workshop_example/bam/workshop_Cer1.bam Cer

./workshop_example/bam/workshop_Cer2.bam Cer

./workshop_example/bam/workshop_Cer3.bam Cer

And then the series of commands I was running are:

majiq build workshop_example/DB.gff3 workshop_example/experiments_info.tsv ./workshop_MAJIQ_build \

--license majiq_license_academic_official.lic --min-experiments 3

majiq psi-coverage ./workshop_MAJIQ_build/splicegraph.zarr/ ./workshop_MAJIQ_psiCoverage/Adr1 ./workshop_MAJIQ_build/workshop_Adr1.sj/

majiq psi-coverage ./workshop_MAJIQ_build/splicegraph.zarr/ ./workshop_MAJIQ_psiCoverage/Adr2 ./workshop_MAJIQ_build/workshop_Adr2.sj/

majiq psi-coverage ./workshop_MAJIQ_build/splicegraph.zarr/ ./workshop_MAJIQ_psiCoverage/Adr3 ./workshop_MAJIQ_build/workshop_Adr3.sj/

majiq psi-coverage ./workshop_MAJIQ_build/splicegraph.zarr/ ./workshop_MAJIQ_psiCoverage/Cer1 ./workshop_MAJIQ_build/workshop_Cer1.sj/

majiq psi-coverage ./workshop_MAJIQ_build/splicegraph.zarr/ ./workshop_MAJIQ_psiCoverage/Cer2 ./workshop_MAJIQ_build/workshop_Cer2.sj/

majiq psi-coverage ./workshop_MAJIQ_build/splicegraph.zarr/ ./workshop_MAJIQ_psiCoverage/Cer3 ./workshop_MAJIQ_build/workshop_Cer3.sj/

majiq deltapsi \

-psi1 workshop_MAJIQ_psiCoverage/Adr1/ workshop_MAJIQ_psiCoverage/Adr2/ workshop_MAJIQ_psiCoverage/Adr3/ \

-psi2 workshop_MAJIQ_psiCoverage/Cer1/ workshop_MAJIQ_psiCoverage/Cer2/ workshop_MAJIQ_psiCoverage/Cer3/ \

--names Adr Cer \

--output-tsv workshop_deltapsi.tsv \

--min-experiments 3 \

--splicegraph workshop_MAJIQ_build/splicegraph.zarr/

majiq build and majiq psi-coverage work well and run successfully. But when I ran majiq deltapsi it gave me the error I shared previously.

Thanks,

Wanru

[San Jewell]

Hi Wanru,

I apologize for the long reply on this, things have been active around the lab and I have also had my fair share of sickness. x3

Anyway I was able to reproduce this issue at last, and strangely it seems to occur when only --output-tsv is specified without also specifying --output-voila . This is certainly not expected behavior, but you can work around it now by also specifying --output-voila to some path until I am able to create a minor version bump and fix. It seems I missed this corner case in testing. Thank you for reporting it.

-San