TensorQTL - trans error

[m-mews]

I'm trying to run the trans analysis using the TensorQTL.ipynb, but I'm receiving the below error. Any thoughts? Thank you in advance!

sos run xqtl-pipeline/code/association_scan/TensorQTL/TensorQTL.ipynb trans \ --genotype-file nhw_239/geno/geno_final/nhw_chr_1_22.no_rel.filtered.bed \ --phenotype-file nhw_239/mol_phe/nhw_gene_tpm_prep_v2.low_expression_filtered.outlier_removed.tmm.expression.bed.gz \ --covariate-file nhw_239/cov/nhw_no_rel.rnaseqc.low_expression_filtered.outlier_removed.tmm.expression.cov_pca.resid.PEER.cov.gz \ --cwd nhw_239/tensorqtl/trans \ --name nhw_no_rel \ --MAC 5 \ --region-list ad_trans_gene_regions.tsv \ --region-name gene_id \ --container xqtl-pipeline/container/singularity/TensorQTL.sif

INFO: Running trans_1: ERROR: trans_1 (id=6f075eae72e9a68e) returns an error. ERROR: [trans_1]: [0]: Executing script in Singularity returns an error (exitcode=1, stderr=/mnt/pan/Data14/metabrain_lasso/rna_seq_norm/nhw_239/tensorqtl/trans/nhw_gene_tpm_prep_v2.low_expression_filtered.outlier_removed.tmm.expression.norminal.trans_long_table.txt.stderr, stdout=/mnt/pan/Data14/metabrain_lasso/rna_seq_norm/nhw_239/tensorqtl/trans/nhw_gene_tpm_prep_v2.low_expression_filtered.outlier_removed.tmm.expression.norminal.trans_long_table.txt.stdout). The script has been saved to /home/mxm1368/.sos/19d2b6ee7290e278//home/mxm1368/.sos/19d2b6ee7290e278.To reproduce the error please run: singularity exec xqtl-pipeline/container/singularity/TensorQTL.sif python /home/mxm1368/.sos/19d2b6ee7290e278/singularity_run_12632.py [trans]: Exits with 1 pending step (trans_2)

singularity exec xqtl-pipeline/container/singularity/TensorQTL.sif python /home/mxm1368/.sos/19d2b6ee7290e278/singularity_run_12632.py Warning: 'rfunc' cannot be imported. R and the 'rpy2' Python package are needed. Mapping files: 100%|██████████████████████████████| 3/3 [01:30<00:00, 30.18s/it] trans-QTL mapping

• 236 samples
• 82 phenotypes
• 44 covariates
• 67510012 variants processing batch 6752/6752 elapsed time: 0.80 min
• 9939322 variants passed MAF >= 0.01 filtering done. trans-QTL mapping (permutations)
• 236 samples
• 44 covariates
• 67510012 variants processing batch 6752/6752 Traceback (most recent call last): File "/home/mxm1368/.sos/19d2b6ee7290e278/singularity_run_12632.py", line 33, in perm_df = trans.map_permutations(genotype_df, covariates_df, batch_size=10000, File "/opt/conda/lib/python3.8/site-packages/tensorqtl/trans.py", line 408, in mappermutations m, = r2_t.max(0) IndexError: max(): Expected reduction dim 0 to have non-zero size.

[gaow]

@m-mews we cannot reproduce it on our end, and we actually don't need permutation testing in our current pipeline using fine-mapping to follow up on marginal assocaitions. I will close the ticket for now. We will release more detailed trans workflow recommendation when i get a chance to document that with inputs from @grennfp . Thanks for your patience!