hybpiper assemble only BWA accomplished, but spades may not be performed

[DL-Lin]

Hi, sorry to bother~

When I use Hybpiper to run my data, i only can get a result folder "D001" include some file show like photo:

My running script is as follows:
hybpiper assemble -t_dna Asparagus-38.fasta -r D001_*.fq.gz --bwa --prefix D001

and i am sure my "target file", that is Asparagus-38.fasta, compliance example "target source and the protein ID, separated by a hyphen", show like:

Asparagus-atp1
ATGGAATTCTCTCCCAGAGCTGCGGAACTCACGACTCTATTAGAAAGTAGAATGACCAACTTTTACACGAATTTTCAAGTGGATGAG.......

I guess my run might not be complete, at least not with spades!!!

but i can complete execution using test_dataset.

so, i don't know what happend of my hybpiper!!

Please let me know if you need any more info, and thanks in advance for your time!!

[chrisjackson-pellicle]

Hi @DL-Lin,

Firstly, the snippet from your target file contains an error, as your fasta header does not begin with a > character. I'm guessing that this is a typo, though, or at least some of your target file headers are formatted correctly, as BWA would not have produced the D001.bam file otherwise.

Can you upload the D001_hybpiper_assemble_2023-12-09-90.log file, please?

Also, do you get any errors printed to the terminal screen when you run the HybPiper command for your data?

Cheers,

Chris

[DL-Lin]

Hi Chris:
Thank you for your busy reply, maybe i have found the reason for the above problem, my run not be complete, because i run my command in the computer cluster with the following parameters:
"#PBS -N iq
#PBS -l nodes=1:ppn=2
#PBS -l walltime=48:00:00
#PBS -q short
#PBS -V
#PBS -S /bin/bash
cd $PBS_O_WORKDIR"

may be run out of disk space!!

when i exchange the parameters:
"#PBS -S /bin/bash
#PBS -N Hybpiper_1
#PBS -l nodes=1:ppn=8
#PBS -l walltime=300:00:00
#PBS -q bigmem"

i can get a result folder "D001" including each gene folder show like photo:

and in each gene folder, i get two files, like this :

but, a new error happend in D001_hybpiper_assemble_2023-12-14-20_04_47.log
see:
D001_hybpiper_assemble_2023-12-14-20_04_47.log

"2023-12-14 20:19:35,385 - assemble.py - hybpiper.assemble - assemble - ERROR - ERROR: No genes had assembled contigs! Exiting!
"
so, why?

Looking forward to your reply, my dear Chris!!!

[chrisjackson-pellicle]

From the log file you uploaded, I can see that SPAdes failed to assemble contigs for all 38 genes in your target file. This might be because not enough reads mapped to your genes of interest, or it might indicate a problem running SPAdes on your system. Can you upload the spades.log file from your sample directory?

Cheers,

Chris

[DL-Lin]

Of course, it's my pleasure!!!!
spades.log

[chrisjackson-pellicle]

Ah, yep, you can see in your spades.log file that SPAdes is exiting with an error for each gene assembly. The issue is:

/datapool/lizhenyu/gaozhiyuan/SOFTWARE/miniconda3/envs/hybpiper/bin/spades-core: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /datapool/lizhenyu/gaozhiyuan/SOFTWARE/miniconda3/envs/hybpiper/bin/spades-core)

It looks like the conda SPAdes package requires at least version 2.14 of GLIBC (the GNU C Library), but the highest version for the system you're running HybPiper on is 2.12 (you can see this in the spades.log file where it states "OS: Linux-2.6.32-642.el6.x86_64-x86_64-with-glibc2.12").

Unfortunately I don't have a solution as far as the HybPiper package goes. GLIBC_2.12, and indeed GLIBC_2.14, are very old (see https://sourceware.org/glibc/wiki/Glibc%20Timeline) - can you ask your systems administrator to update the GLIBC library, if possible?

Cheers,

Chris

[perezii]

Hi Chris,
I have kind of the same problem (SPAdes did not perform). I saw the spades.log and found this:

System information:
SPAdes version: 3.15.5
Python version: 3.11.7
OS: Linux-4.18.0-477.15.1.el8_8.x86_64-x86_64-with-glibc2.28

The glibc is 2.28 and I used Hybpiper v2.1.5

Thanks
Miguel

[perezii]

Here is my log
spades_2.log

[chrisjackson-pellicle]

Hi @perezii,

Your issue is different to the SPAdes / GLIBC problem described above. Can you open a new issue, and upload your full spades.log file, along with your sample log file (11L_aguana_hybpiper_assemble_<date_time>.log) from your 11L_aguana sample directory?

Cheers,

Chris