search

chhylp123 / hifiasm

Hifiasm: a haplotype-resolved assembler for accurate Hifi reads
MIT License
547 stars 87 forks source link

Error Message reads: "Killed" #200

Open smvelazquez opened 3 years ago

smvelazquez commented 3 years ago

I'm currently working on assembling HiFi reads for a highly heterozygous species. When I run the following command:

hifiasm -o <name>.asm -t49 -f39 read.1.fastq.gz read.2.fastq.gz

I receive the following error message:

Killed.

For context, I am running on a c5.9xlarge AWS instance that has plenty of space and RAM, so theoretically power should not be an issue. The command very quickly times out (5 minutes or less) and crashed my connection to the instance. This is also being run on a screen session on the instance so the crashed connection shouldn't cause an issue, but it has been a pattern that when the connection crashes the hifiasm command also crashed. The first file is 14 GB while the second is 24 GB.

This has also been true for another sample I have been trying to create a phased assembly in which the files are also large, Hi-C reads are ~60 GB and HiFi reads (4 fastqs) ~15 GB.

chhylp123 commented 3 years ago

Could you please show the log file? Did you samples have weird k-mer plots (see: https://hifiasm.readthedocs.io/en/latest/faq.html#why-does-hifiasm-stuck-or-crash)?

smvelazquez commented 3 years ago

No files are produced after the crash. The directory remains empty aside from the input files.

chhylp123 commented 3 years ago

Could you please share one of them with us as an example? Killed should be reported by linux system instead of hifiasm.

chhylp123 commented 3 years ago

Or you can get k-mer plot by yourself with tools like jellyfish.

smvelazquez commented 3 years ago

Unfortunately I can't share the data. Is there a software or something you recommend so I can check the quality of the HiFi reads? I can report back the results here.

chhylp123 commented 3 years ago

You can have a try with jellyfish. Thanks.