Abyss - error Makefile:667

[desmodus1984]

Please report

• [ ] version of ABySS with abyss-pe version
• [ ] (abyss) [juaguila@u01 abyss-763]$ abyss-pe version bash: line 0: test: -le: unary operator expected abyss-pe (ABySS) 2.3.7 Written by Shaun Jackman and Anthony Raymond.

Copyright 2012 Canada's Michael Smith Genome Science Centre

• [ ] distribution of Linux with lsb_release -d

(abyss) [juaguila@u01 abyss-763]$ lsb_release -d bash: lsb_release: command not found (base) [juaguila@u05 ~]$ cat /etc/os-release NAME="CentOS Linux" VERSION="7 (Core)" ID="centos" ID_LIKE="rhel fedora" VERSION_ID="7" PRETTY_NAME="CentOS Linux 7 (Core)" ANSI_COLOR="0;31" CPE_NAME="cpe:/o:centos:centos:7" HOME_URL="https://www.centos.org/" BUG_REPORT_URL="https://bugs.centos.org/"

CENTOS_MANTISBT_PROJECT="CentOS-7" CENTOS_MANTISBT_PROJECT_VERSION="7" REDHAT_SUPPORT_PRODUCT="centos" REDHAT_SUPPORT_PRODUCT_VERSION="7"

Assembly error

• [ ] complete abyss-pe command line

• [ ] (after loading conda and abyss environment) nohup abyss-pe name=763 k=96 B=10G G=200M j=10 in='../763-t1.fq.gz ../763-t2.fq.gz'

• [ ] last 20 lines of the output of abyss-pe tail -n 20 nohup.out Unique path: 37072 No possible paths: 12082 No valid paths: 8 Repetitive: 0 Multiple valid paths: 347 Too many solutions: 0 Too complex: 1

The minimum number of pairs in a distance estimate is 10. The minimum number of pairs used in a path is 10. abyss-index --fai 763-3.fa Reading 763-3.fa'... Writing 763-3.fa.fai'... abyss-index --fai 763-4.fa Reading 763-4.fa'... Writing 763-4.fa.fai'... cat 763-3.fa.fai 763-4.fa.fai
| MergePaths -j10 -k96 -s1000 -G200M -o 763-4.path2 - 763-4.path1 MergePaths: invalid option: `-G200M' make: *** [/home/juaguila/.conda/envs/abyss/bin/abyss-pe.Makefile:667: 763-4.path2] Error 1

• [ ] number of sequenced bases * GB
• [ ] estimated genome size and ploidy diploid: 185 MB by JellyFish
• [ ] estimated sequencing depth of coverage about 50 X

I installed Abyss in conda which worked fine, but then it failed during assembly. I ran it in a server as shown above which has 54 cores and 500GB of RAM.

Any reason why it did fail?

Thanks;

[lcoombe]

Hi @desmodus1984,

Could you try a fresh ABySS run, specifying G=200000000 or G=200e6 instead? I believe that it is failing due to using the M unit instead of providing the raw genome size number.

Thanks for your interest in ABySS! Lauren

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your interest in ABySS!

[desmodus1984]

Hi, I did try the suggestion and it worked. I wanted to ask a question. Do you think that assembly will benefit from merging paired-end reads and use them as single-end reads? or it won't make much difference and just use straight trimmed paired-end reads as forward/reverse reads in a Bloom filter.

Thanks.

[lcoombe]

Hi @desmodus1984,

Glad that tweak worked!

In terms of read merging, it is often something that we do try on our end when doing parameter sweeps with ABySS. The benefit of read merging is that you can generally increase the k-mer size, which can benefit the assembly in a lot of cases. Of course this is dataset specific, so there aren't any guarantees.

If you try it, the important thing is to ensure that you still provide the paired-end unmerged reads to ABySS for the contig and scaffolding stages. To do that, you'd use se= for your single-end or merged reads, and pe= for your paired-end reads. I'd suggest doing a dry-run -n first to ensure that the single-end reads are being used in the first stage (unitig stage, up to -3 output files), and the paired-end reads are being used in the subsequent contig and scaffold stages.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your interest in ABySS!