SCENICprotocol
SCENICprotocol copied to clipboard
aertslab
How long for pyscenic ctx to run?
Hello,
I'm currently running the Docker version of pyscenic (0.12.1) on HPC (Fedora). All was OK until I hit the pyscenic ctx command. My command is below:
pyscenic ctx \ ./expr_mat.adjacencies.tsv \ software/SCENIC/resources/databases/new/homo_sapiens/hg38/refseq_r80/mc_v10_clust/gene_based/hg38_10kbp_up_10kbp_down_full_tx_v10_clust.genes_vs_motifs.rankings.feather \ software/SCENIC/resources/databases/new/homo_sapiens/hg38/refseq_r80/mc_v10_clust/gene_based/hg38_500bp_up_100bp_down_full_tx_v10_clust.genes_vs_motifs.rankings.feather \ --annotations_fname software/SCENIC/resources/motif2tf/motifs-v10nr_clust-nr.hgnc-m0.001-o0.0.tbl \ --expression_mtx_fname ./expr_mat_filtered.tsv \ --mode "custom_multiprocessing" \ --output ./regulons_mask_dropouts.csv \ --mask_dropouts
So far, I've tried running it multiple times and it always goes quickly until the "pyscenic.prune" step where I will get the following messages and then it freezes indefinitely (so far we are at 2 days for one of my attempts):
`2023-07-18 16:01:10,759 - pyscenic.prune - INFO - Worker hg38__refseq-r80__10kb_up_and_down_tss.mc9nr.genes_vs_motifs.rankings(14): All regulons derived.
2023-07-18 16:01:10,759 - pyscenic.prune - INFO - Worker hg38__refseq-r80__10kb_up_and_down_tss.mc9nr.genes_vs_motifs.rankings(14): All regulons derived.
2023-07-18 16:01:11,050 - pyscenic.prune - INFO - Worker hg38__refseq-r80__10kb_up_and_down_tss.mc9nr.genes_vs_motifs.rankings(14): Done.
2023-07-18 16:01:11,050 - pyscenic.prune - INFO - Worker hg38__refseq-r80__10kb_up_and_down_tss.mc9nr.genes_vs_motifs.rankings(14): Done. ` I get these messages for presumably all 32 workers? It will give this message for numbers 1-16.
What can I do to get past this infinitely-running part of pyscenic ctx? Or does it just take a long time? For context, I have about 7k cells, but I'm using a SMART-seq-based method, so more genes than 10x.
