Nextpolish unfinished about nextpolish HOT 3 OPEN

1. The max memory required by NextPolish is depending on the value of -p, each sub-process required 2-5Gb RAM depending on the mapped data depth. So 64 Gb of Ram is enough if you remove -a options, which will automatically adjust -p.
2. Some of sub-processes are crashed (usually caused by insufficient memory), which blocked the main process. This is a bug of multiprocess module in python. so you should kill the main task, and just rerun the main task. It it will skip corrected seqs and continue running.
3. It is better to update NexPolish to the latest version and try again. The polishing step is very fast, it should be finished in few minutes depending on your genome size and -p.

from nextpolish.

Indeed, delete the -p -a option make it work again. I relaunch the the command and it works.

My corrected file, pb.asm.nextpolish1.fa, has multiple time the same deflines .

grep scaffold_300 pb.asm.nextpolish1.fa

scaffold_300_pilon_pilon_subseq_1:49068_obj 49236
scaffold_300_pilon_pilon_subseq_1:49068_obj 49236
scaffold_300_pilon_pilon_subseq_1:49068_obj 49236

Is it because I run the pipeline multiple times, can I just keep one of the sequences ?

Thanks,
Luc

from nextpolish.

In theory, NextPolish will skip polished contigs, so it should not have duplicate IDs, I need more time to debug it. By now, it is more safer to remove pb.asm.nextpolish1.fa and rerun this step, it should be finished in 10 -30 minutes if mapped depth < 100.

from nextpolish.