HiC-/PoreC consensus failed consistently at 7-consensus stage with both Hi-C and Pore-C/CiFi data

[txfc]

I have been trying to run Verkko (v2.2) to achieve a T2T assembly for a bird genome (bird, estimated genome size ~1.1 Gb, heterozygosity ~0.6%). I have PacBio HiFi data (~120 Gb), ONT ultra-long reads (R10.4, N50 > 70 kb), and both Arima Hi-C (100 Gb) and PacBio CiFi (50 Gb) data. However, the pipeline consistently fails at the 7-consensus stage with the error ERROR: HiC-/PoreC consensus failed, regardless of whether I use Hi-C or Pore-C/CiFi data for phasing.

Environment

Verkko version: 2.2 (installed via conda)

Resource: 256 CPU cores, 1000 GB RAM

Run Verkko with Hi‑C phasing:

verkko -d asm_hic \
    --hifi hifi_reads.fastq.gz \
    --nano ont_ul.fastq.gz \
    --hic1 hic_R1.fastq.gz --hic2 hic_R2.fastq.gz \
    --local-cpus 128 

Run Verkko with Pore‑C phasing (using the same HiFi and ONT data):

verkko -d asm_porec \
    --hifi hifi_reads.fastq.gz \
    --nano ont_ul.fastq.gz \
    --porec cifi.fastq.gz \
    --local-cpus 128 

Both runs progressed through the early stages (0‑correction~ 8-hicPipeline, etc.) but eventually failed with the same error at the 7-consensus stage.

This is error messages from hic data，and it is similar with the error from cifi data.

Finished job 48.
25 of 28 steps (89%) done
Exiting because a job execution failed. Look above for error message
Complete log: .snakemake/log/2026-06-07T075943.571731.snakemake.log
ERROR: HiC-/PoreC consensus failed, look above for error message
Verkko logging for failed commands: ./packages/part011.sh > ../7-consensus/packages/part011.err 2>&1
7-consensus/packages/part013.err:
	    673    127859      28         3    0.20x        0    0.00x        13    0.99x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 5 tigs and 1 singleton.
	-- 
7-consensus/packages/part015.err:
	    868    230666     198        53    2.85x        0    0.00x        38    2.59x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 8 tigs and 0 singletons.
	-- 
7-consensus/packages/part018.err:
	    782     80622       9         1    0.12x        0    0.00x         4    0.54x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 8 tigs and 1 singleton.
	-- 
7-consensus/packages/part016.err:
	    860    171645     196       105    5.79x        0    0.00x        55    3.53x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 7 tigs and 0 singletons.
	-- 
7-consensus/packages/part017.err:
	    503    142854      87        58    4.18x        0    0.00x        17    1.48x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 7 tigs and 0 singletons.
	-- 
7-consensus/packages/part012.err:
	    563    163509      23         3    0.18x        0    0.00x        13    0.78x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 5 tigs and 0 singletons.
	-- 
7-consensus/packages/part002.err:
	     31  88208949  430962    306226   41.55x    46365    5.04x     47487    8.43x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 1 tig and 0 singletons.
	-- 
7-consensus/packages/part005.err:
	    166  71738634  354861    249515   41.66x    37711    5.04x     38198    8.34x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 1 tig and 0 singletons.
	-- 
7-consensus/packages/part020.err:
	    676     84017       8         2    0.22x        0    0.00x         4    0.53x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 9 tigs and 1 singleton.
	-- 
7-consensus/packages/part009.err:
	    740    250155     230       128    5.44x        0    0.00x        43    3.13x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 6 tigs and 0 singletons.
	-- 
7-consensus/packages/part010.err:
	    517   2225072    9139      7247   37.69x        0    0.00x      1148    7.90x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 5 tigs and 1 singleton.
	-- 
7-consensus/packages/part008.err:
	    955   5651981   25300     20356   41.80x        0    0.00x      2970    8.15x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 5 tigs and 0 singletons.
	-- 
7-consensus/packages/part004.err:
	      2  77628029  387621    268192   41.50x    45623    5.67x     41939    8.48x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 1 tig and 0 singletons.
	-- 
7-consensus/packages/part006.err:
	    809    286677     101        36    1.22x        0    0.00x        46    1.80x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 4 tigs and 0 singletons.
	-- 
7-consensus/packages/part022.err:
	    509    774860    1778      1281   19.17x        0    0.00x       381    6.76x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 10 tigs and 0 singletons.
	-- 
7-consensus/packages/part003.err:
	    120  77791749  387936    268984   41.57x    45533    5.65x     41693    8.42x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 1 tig and 0 singletons.
	-- 
7-consensus/packages/part014.err:
	    819    158888      20         3    0.18x        0    0.00x        14    0.88x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 6 tigs and 0 singletons.
	-- 
7-consensus/packages/part019.err:
	    811     20377       5         1    0.49x        0    0.00x         4    2.16x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 8 tigs and 0 singletons.
	-- 
7-consensus/packages/part024.err:
	    972     38073      16         4    0.93x        0    0.00x         9    2.54x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 815 tigs and 11 singletons.
	-- 
7-consensus/packages/part007.err:
	    971     73195       4         2    0.26x        0    0.00x         1    0.20x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 5 tigs and 0 singletons.
	-- 
7-consensus/packages/part021.err:
	    846     66788       9         0    0.00x        0    0.00x         5    0.67x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 9 tigs and 0 singletons.
	-- 
7-consensus/packages/part011.err:
	-rw-rw-r-- 1 duanying duanying        769 Jun  7 13:52 part001.err
	-rw-rw-r-- 1 duanying duanying  130982201 Jun  7 13:52 part001.fasta
	-rw-rw-r-- 1 duanying duanying   77672448 Jun  7 14:37 part011.fasta.WORKING
	-rw-rw-r-- 1 duanying duanying 1467628517 Jun  7 14:41 part011.bam00000040.bam
	-rw-rw-r-- 1 duanying duanying       1854 Jun  7 14:41 part011.err
7-consensus/packages/part001.err:
	    156  91693647  449880    318366   41.56x    48245    5.05x     49350    8.43x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 1 tig and 0 singletons.
	-- 
7-consensus/packages/part023.err:
	    777    132399      26         8    0.57x        0    0.00x        14    1.13x
	------- --------- -------  -------- -------- -------- --------  -------- --------
	--
	-- Processed 25 tigs and 0 singletons.
	-- 

Thank you for developing Verkko and for your time.

[skoren]

It's possible this is a bug that was fixed post-release. If you can share the 8-hicPipeline/final*/7-consensus/packages/part011.err file and the 8-hicPipeline/final*/7-consensus/packages/part011.cnspack with us we can try to reproduce the error locally. You can upload the the Google Drive here: https://drive.google.com/drive/folders/1OGflPHE43VKxaCKwrmMvIE8DF1iy6ZKj?usp=sharing

[txfc]

part011.txt
Ok I have uploaded this error file and I will try to upload this part011.cnspack file.

[txfc]

https://drive.google.com/file/d/1iUL5cjHmIPcHXkOJ96DE_02RmO-NEam_/view?usp=sharing I have uploaded the file here.

[skoren]

Thanks, the link for the err file looks like it isn't working, could you upload it again along with the part011.sh file?

[txfc]

https://drive.google.com/file/d/1iUL5cjHmIPcHXkOJ96DE_02RmO-NEam_/view?usp=sharing

[txfc]

Unfortunately, I re-tried the ont data with Herro's correction, which contained both 9.4 and 10.4 inputs, and reported a similar error. This is the error document

Exiting because a job execution failed. Look above for error message
Complete log: .snakemake/log/2026-06-10T115619.287705.snakemake.log
Verkko logging for failed commands:     input: /home/duanying/refgenome_assembly/ont_data/batch2-N50_100K/batch2_herro_10kb.fasta.gz, /home/duanying/refgenome_assembly/ont_data/batch1-ont-100k/corrected_filtered.fasta.gz, 2-processGraph/unitig-unrolled-hifi-resolved.gfa, 3-align/alns-ont.gaf, 3-align/splitONT.finished
./packages/part019.sh > ../7-consensus/packages/part019.err 2>&1

This is part019.err

-- Computing consensus for b=1 to e=4294967295 with errorRate 0.0500 (max 0.1000) and minimum overlap 500
--
generateTemplateStitch()-- Warning: failed to build tig 1 template at error rate 0.033333, trying 0.041667
generateTemplateStitch()-- Warning: failed to build tig 1 template at error rate 0.041667, trying 0.052083
generateTemplateStitch()-- Warning: failed to build tig 1 template at error rate 0.052083, trying 0.065104
generateTemplateStitch()-- Warning: failed to build tig 1 template at error rate 0.065104, trying 0.081380
generateTemplateStitch()-- Warning: failed to build tig 1 template at error rate 0.081380, trying 0.101725
utgcns: utgcns/unitigConsensus.C:524: char* unitigConsensus::generateTemplateStitch(): Assertion `ii >= _numReads' failed.

Failed with 'Aborted'; backtrace (libbacktrace):

Failed with 'Segmentation fault'; backtrace (libbacktrace):
Segmentation fault (core dumped)

Consensus did not finish successfully, exit code 0.

Could it be that there is something wrong with my ont data?
This is my code

verkko 
-d ./ 
--hifi "/home/refgenome_assembly/hifi_data/batch1_hifi.fastq.gz" "/home/refgenome_assembly/hifi_data//batch2.fq.gz" 
--nano "/home/refgenome_assembly/ont_data/batch2_herro_10kb.fasta.gz" "/home/duanying/refgenome_assembly/ont_data/corrected_filtered.fasta.gz"  
--hap-kmers "/home//refgenome_assembly/paternal_compress.k31.hapmer.meryl" "/home//refgenome_assembly/maternal_compress.k31.hapmer.meryl" trio  
--local-cpus 128 

HIFI110x and ont 30x（9.4） + 30x（10.4）

[skoren]

Ah I missed you're on version 2.2, you're likely hitting an old bug mentioned in this issue #347. I expect it is fixed in 2.3.2 but that is not backwards compatible with your version. You could try the workaround suggested there (increasing error rate) and see if it works for your assembly.

I'd like to confirm it is fixed in 2.3.2 but I still only see the cnspack in the upload, I need the sh files as well (at least what it says for -E), can you share that for the original part011.cnspack?

I wouldn't suggest including ONT R9 data into the corrected inputs, it is much lower quality than R10 and likely would introduce noise rather than helping.

[txfc]

Sorry, I misread it. My version is 2.3.2. I only used the uncorrected 10.4 ont data when running hic and Porec modes.
This is part11.sh.

#!/bin/sh
set -e

if [ '7-consensus/packages/part011.bam' != 'emptyfile' ]; then
   align="-B ../7-consensus/packages/part011.bam"
else
   align="-norealign"
fi

minread=100
if [ "$(expr 4000 - 500)" -gt 0 ]; then
    minread=500
fi

# when generating initial consensus skip bam generation and only 1 round
if [ 'False' = 'True' ]; then
   align="-norealign"
   iter="-I 1"
else
   iter="-I 2"
fi

/home/duanying/miniforge3/envs/verkko/lib/verkko/bin/utgcns \
    -threads 8 \
    -import ../7-consensus/packages/part011.cnspack \
    -A ../7-consensus/packages/part011.fasta.WORKING \
    $iter -C 2 \
    $align \
    -maxcoverage 50 \
    -e  0.05 \
    -em 0.10 \
    -EM 876551 \
    -l $minread \
&& \
mv ../7-consensus/packages/part011.fasta.WORKING ../7-consensus/packages/part011.fasta \
&& \
exit 0

echo ""
echo "Consensus did not finish successfully, exit code $?."

echo ""
echo "Files in current directory:"
ls -ltr

echo ""
echo "Files in packages/:"
ls -ltr packages

exit 1

[skoren]

I confirmed this is a bug and it is fixed in a commit to master: marbl/canu@86d7d3b. If you're able to check out this branch from canu: https://github.com/marbl/canu/tree/consensus-skiplowqual-test and build it locally, that would give you a patched utgcns to complete your asm. You can then copy it into /home/duanying/miniforge3/envs/verkko/lib/verkko/bin/utgcns and resume the assembly. If that doesn't work, I can try to make a patched binary for you to complete the asm.

[majssssa]

Hi ,
Thank you very much for your help. I followed your instructions and checked out the consensus-skiplowqual-test branch of Canu. I successfully built the patched utgcns and replaced the original utgcns in my verkko environment:
/home/majunpeng/miniforge3/envs/verkko/lib/verkko/bin/utgcns
However, after replacing the binary, the assembly still cannot proceed successfully. The error remains when running the consensus step.
Could you please advise what else I should check? If possible, could you provide a patched binary of utgcns compiled for Linux, so I can directly test whether the issue is related to my local compilation environment?
I really appreciate your help.
Best regards,
Junpeng

[skoren]

Strange, is it failing on the same partition or a different one, perhaps there is more than one bug?

I attached a gziped compiled binary version of utgcns which I confirmed ran on the part011.cnspack you shared previously. Test it first on that partitionto make sure it works before running the rest of the assembly.
3dcf846149c188af61d78861c00fc854 utgcns.gz

utgcns.gz