Fail to deconstruct VCF from the PGGB GFA file

[isaamael]

Hi !
I am trying to generate VCF from the GFA produced by the PGGB pipeline.
I attempted to use deconstruct, and everything works fine on smaller GFA files, such as when I only include two species. However, when I include 11 species ,which I actually need, vg only outputs the VCF header. and does not produce any error messages.

Could it be that the task takes too long to run and gets killed by SLURM, or have I made some hidden mistake?

The command I ran:

i=1
vg deconstruct --verbose --path-prefix solyc --all-snarls --path-traversals --threads 10 --untangle-travs --contig-only-ref \
$pggbdir/sly${i}.fa.out/sly${i}.fa.bf3285f.417fcdf.7fecd6e.smooth.final.gfa > $pggbdir/pan${i}.vcf

It seems that the command runs for a long time without producing any output and is eventually stopped by SLURM?All Log from the SLURM systerm:

Warning [vg deconstruct]: -e is deprecated as it's now on default
slurmstepd: error: *** JOB 17088635 ON comput52 CANCELLED AT 2024-11-14T21:40:23 DUE TO TIME LIMIT ***

And any other vg commands, such as combine and convert, will produce the same error...

I am willing to provide any necessary information and would appreciate your help!

[AndreaGuarracino]

@isaamael, it looks like you're right, vg deconstruct doesn't finish on time and doesn't write anything! Can you try to request more time for the job?

[isaamael]

@isaamael, it looks like you're right, vg deconstruct doesn't finish on time and doesn't write anything! Can you try to request more time for the job?

Hello sir,

Thanks for your suggestions.

In theory, there should be no time limit on the SLURM system I'm using, so I'm not sure why I'm encountering timeout issues. After manually defining the runtime, I generated some low-confidence VCF, which I attribute to the interference caused by the low-quality genome during the GFA construction.

In any case, after filtering the original input genome, I think it would be best to rerun pggb and generate the VCF directly from the pipeline. However, I tried the following command and encountered an error:

pggb -i $pggbdir/sly12.fa \
     -o $pggbdir/sly12.fa.out \
     -p 5000 -l 25000 -p 90 -n 2 -K 19 -F 0.001 -t 20 \
     -k 19 -f 0 -B 10000000 \
     -j 0 -e 0 -G 700,900,1100 -P 1,19,39,3,81,1 -O 0.001 -d 100 -Q Consensus_ \
     --vcf-spec 'solyc#0#12'

[vg::deconstruct] making VCF with reference=solyc#0#12 and delim=# xxxxxxxxxxxxx solyc#0#12 ------------ 0
Error [vg deconstruct]: No specified reference path or prefix found in graph
Command exited with non-zero status 1

My reference is indeed solyc#0#12, and the path in the output GFA is the same. I'm unclear about the correct way to specify --vcf-spec parameter and how to define LEN.

Some beginner questions and would appreciate your help : )

[AndreaGuarracino]

Just specify "... --vcf-spec solyc" and it should work!

…

________________________________ From: Isaac ***@***.***> Sent: Wednesday, November 27, 2024 21:54 To: pangenome/pggb ***@***.***> Cc: Andrea Guarracino ***@***.***>; Comment ***@***.***> Subject: Re: [pangenome/pggb] Fail to deconstruct VCF from the PGGB GFA file (Issue #429) @isaamael<https://github.com/isaamael>, it looks like you're right, vg deconstruct doesn't finish on time and doesn't write anything! Can you try to request more time for the job? Hello sir, Thanks for your suggestions. In theory, there should be no time limit on the SLURM system I'm using, so I'm not sure why I'm encountering timeout issues. After manually defining the runtime, I generated some low-confidence VCF, which I attribute to the interference caused by the low-quality genome during the GFA construction. In any case, after filtering the original input genome, I think it would be best to rerun pggb and generate the VCF directly from the pipeline. However, I tried the following command and encountered an error: pggb -i $pggbdir/sly12.fa \ -o $pggbdir/sly12.fa.out \ -p 5000 -l 25000 -p 90 -n 2 -K 19 -F 0.001 -t 20 \ -k 19 -f 0 -B 10000000 \ -j 0 -e 0 -G 700,900,1100 -P 1,19,39,3,81,1 -O 0.001 -d 100 -Q Consensus_ \ --vcf-spec 'solyc#0#12' [vg::deconstruct] making VCF with reference=solyc#0#12 and delim=# xxxxxxxxxxxxx solyc#0#12 ------------ 0 Error [vg deconstruct]: No specified reference path or prefix found in graph Command exited with non-zero status 1 My reference is indeed solyc#0#12, and the path in the output GFA is the same. I'm unclear about the correct way to specify --vcf-spec parameter and how to define LEN. Some beginner questions and would appreciate your help : ) — Reply to this email directly, view it on GitHub<#429 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AO26XHXBL2Z6ZXU4RN53BJT2C2HWDAVCNFSM6AAAAABSBAERTGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMBVGIYTGNBVGA>. You are receiving this because you commented.Message ID: ***@***.***>

[isaamael]

i will try it !

[isaamael]

Hi sir, @AndreaGuarracino
i try cmd

pggb -i $pggbdir/sly12.fa \
     -o $pggbdir/sly12.fa.out \
     -p 5000 -l 25000 -p 90 -n 10 -K 19 -F 0.001 -t 20 \
     -k 19 -f 0 -B 10000000 \
     -j 0 -e 0 -G 700,900,1100 -P 1,19,39,3,81,1 -O 0.001 -d 100 -Q Consensus_ \
     --vcf-spec solyc

and report an error

[wfmash::align::computeAlignments] aligned 78.50% @ 2.94e+05 bp/s elapsed: 00:02:17:56 remain: 00:00:37:46[E::fai_retrieve] Failed to retrieve block: unexpected end of file
Command terminated by signal 11

Did i make some mistakes ?

[AndreaGuarracino]

Oh no! It seems your FASTA file is corrupted. Can you check that your input FASTA is healthy? Sent from Outlook for Android<https://aka.ms/AAb9ysg>

…

________________________________ From: Isaac ***@***.***> Sent: Saturday, November 30, 2024 3:10:09 AM To: pangenome/pggb ***@***.***> Cc: Andrea Guarracino ***@***.***>; Mention ***@***.***> Subject: Re: [pangenome/pggb] Fail to deconstruct VCF from the PGGB GFA file (Issue #429) Hi sir, @AndreaGuarracino<https://github.com/AndreaGuarracino> i try cmd pggb -i $pggbdir/sly12.fa \ -o $pggbdir/sly12.fa.out \ -p 5000 -l 25000 -p 90 -n 10 -K 19 -F 0.001 -t 20 \ -k 19 -f 0 -B 10000000 \ -j 0 -e 0 -G 700,900,1100 -P 1,19,39,3,81,1 -O 0.001 -d 100 -Q Consensus_ \ --vcf-spec solyc and report an error [wfmash::align::computeAlignments] aligned 78.50% @ 2.94e+05 bp/s elapsed: 00:02:17:56 remain: 00:00:37:46[E::fai_retrieve] Failed to retrieve block: unexpected end of file Command terminated by signal 11 Did i make some mistakes ? — Reply to this email directly, view it on GitHub<#429 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AO26XHUF463GI72J6BGUR6T2DF6HDAVCNFSM6AAAAABSBAERTGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMBYHA4TKNRVGE>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[isaamael]

yes! indeed
I denove generated the FASTA and performed pggb.
Now, GFA, VCF, and I are all as good as expected :)
Sincere thanks, thanks for your help!

Additionally, do the numbers in the GT field of the VCF point to the paths of AT?

Oh no! It seems your FASTA file is corrupted. Can you check that your input FASTA is healthy? Sent from Outlook for Androidhttps://aka.ms/AAb9ysg
…
________________________________ From: Isaac @.> Sent: Saturday, November 30, 2024 3:10:09 AM To: pangenome/pggb @.> Cc: Andrea Guarracino @.>; Mention @.> Subject: Re: [pangenome/pggb] Fail to deconstruct VCF from the PGGB GFA file (Issue #429) Hi sir, @AndreaGuarracinohttps://github.com/AndreaGuarracino i try cmd pggb -i $pggbdir/sly12.fa \ -o $pggbdir/sly12.fa.out \ -p 5000 -l 25000 -p 90 -n 10 -K 19 -F 0.001 -t 20 \ -k 19 -f 0 -B 10000000 \ -j 0 -e 0 -G 700,900,1100 -P 1,19,39,3,81,1 -O 0.001 -d 100 -Q Consensus_ \ --vcf-spec solyc and report an error [wfmash::align::computeAlignments] aligned 78.50% @ 2.94e+05 bp/s elapsed: 00:02:17:56 remain: 00:00:37:46[E::fai_retrieve] Failed to retrieve block: unexpected end of file Command terminated by signal 11 Did i make some mistakes ? — Reply to this email directly, view it on GitHub<#429 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO26XHUF463GI72J6BGUR6T2DF6HDAVCNFSM6AAAAABSBAERTGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMBYHA4TKNRVGE. You are receiving this because you were mentioned.Message ID: @.***>