Open Zick007 opened 4 years ago
I have the same issue here. However the step is running for more than a day. I have even tried the parameter -nsl
but it seems that the pipeline have ignored it.
Same here, 22 minutes running on 11 genomes using 14 threads. Maybe skipping step 6 is an option?
"====== time for step06: 22.01 minutes (1320.48 seconds)"
Hi,
I runned the command-line ./panX.py -fn /.data/MyData -sl T_species
Everything seemed good till step 6 I think... Now this step is running for 10 minutes. I only have 5 gbk files to compare in MyData. Is it normal? Should I let it continue or it's a bug?
I already checked and I've correctly installed all dependencies.
Also, I encountered the same problem when I launched the test.sh script...
Thank you!