Synteny analysis for large genomes

[desmodus1984]

Hello,
I am interested in using ntSynt to check for synteny of genomes of marine mammals, which are about 2.5 GB each.
I read that is recommended to use Mash to check for divergence, and I used it with the regular command:
I sketched the genomes
mash sketch genome1.fna
mash sketch genome2.fna
and the I calculated the distance for each of the 4 genomes
mash dist genome1.fna.msh genome2.fna.msh

I read that one could adjust some parameters, and I was wondering about if what I did is enough or if I should modify any parameters for doing the divergence estimation.
Most of the genomes, 3, were scaffolded using Hi-C.

Thanks;

[lcoombe]

Hi @desmodus1984,

Yes, that is right, we recommend using Mash to estimate the divergence, and then setting that parameter to the max (%) estimated. For an initial run, this is sufficient - setting the input divergence will auto-set the parameters.
Once you have a sucsessful run, you could go back and adjust some parameters if you wanted to tweak things like the indel threshold between blocks, collinear merge length between blocks, etc. But most of the time, I will just leave the presets as-is.

I hope that makes sense - thank you for your interest in ntSynt!
Lauren

[github-actions]

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 ntSynt!

[sanyalab]

Hello Lauren,

I am similarly trying to use nySynt for insect genomes. But I run into an error. The run_demo.sh executed properly. I have ntSynt installed using conda. I performed the following steps.

1. Calculated mash distance and I get the output like below
   genome1.fna genome2.fna 0.2562 0 225/1000. Although here I do not understand whether -d should be 25% or 0.25%. I used both in separate runs but ended up getting errors in both.

2. My command for ntSynt: ntSynt --fastas_list genome.path -d 0.25 (The genome.path contains paths to 8 insect genomes). The pipeline is failing with following error

2026-02-05 08:17:43.712261 : Finding synteny blocks
Error - no paths found. Try adjusting the specified k/w parameters.
RuleException:
CalledProcessError in file "/GenomeQCEnv/bin/ntsynt_run_pipeline.smk", line 102:
Command 'set -euo pipefail;   python3 /GenomeQCEnv/bin/ntsynt_run.py genome1_chr.fa.k24.w1000.tsv genome2_chr.fa.k24.w1000.tsv genome3_chr.fa.k24.w1000.tsv genome4_chr.fa.k24.w1000.tsv genome5_chr.fa.k24.w1000.tsv genome6_chr.fa.k24.w1000.tsv genome7_chr.fa.k24.w1000.tsv genome8_chr.fa.k24.w1000.tsv -k 24 -w 1000 --w-rounds 100 10 -p ntSynt.k24.w1000 --bp 10000 --collinear-merge 10000 -z 500 --common ntSynt.k24.w1000.common.bf  --simplify-graph              --btllib_t 12  --fastas /SYNTENY/genome1_chr.fa /SYNTENY/genome2_chr.fa /SYNTENY/genome3_chr.fa /SYNTENY/genome4_chr.fa /SYNTENY/genome5_chr.fa /SYNTENY/genome6_chr.fa /SYNTENY/genome7_chr.fa /SYNTENY/genome8_chr.fa' returned non-zero exit status 1.
[Thu Feb  5 08:17:43 2026]
Error in rule ntsynt_synteny:
    message: None
    jobid: 10
    input: genome1_chr.fa.k24.w1000.tsv, genome2_chr.fa.k24.w1000.tsv, genome3_chr.fa.k24.w1000.tsv, genome4_chr.fa.k24.w1000.tsv, genome5_chr.fa.k24.w1000.tsv, genome6_chr.fa.k24.w1000.tsv, genome7_chr.fa.k24.w1000.tsv, genome8_chr.fa.k24.w1000.tsv, ntSynt.k24.w1000.common.bf, genome1_chr.fa.fai, genome2_chr.fa.fai, genome3_chr.fa.fai, genome4_chr.fa.fai, genome5_chr.fa.fai, genome6_chr.fa.fai, genome7_chr.fa.fai, genome8_chr.fa.fai, /SYNTENY/genome1_chr.fa, /SYNTENY/genome2_chr.fa, /SYNTENY/genome3_chr.fa, /SYNTENY/genome4_chr.fa, /SYNTENY/genome5_chr.fa, /SYNTENY/genome6_chr.fa, /SYNTENY/genome7_chr.fa, /SYNTENY/genome8_chr.fa
    output: ntSynt.k24.w1000.synteny_blocks.tsv
    shell:
         python3 /GenomeQCEnv/bin/ntsynt_run.py genome1_chr.fa.k24.w1000.tsv genome2_chr.fa.k24.w1000.tsv genome3_chr.fa.k24.w1000.tsv genome4_chr.fa.k24.w1000.tsv genome5_chr.fa.k24.w1000.tsv genome6_chr.fa.k24.w1000.tsv genome7_chr.fa.k24.w1000.tsv genome8_chr.fa.k24.w1000.tsv -k 24 -w 1000 --w-rounds 100 10 -p ntSynt.k24.w1000 --bp 10000 --collinear-merge 10000 -z 500 --common ntSynt.k24.w1000.common.bf  --simplify-graph              --btllib_t 12  --fastas /SYNTENY/genome1_chr.fa /SYNTENY/genome2_chr.fa /SYNTENY/genome3_chr.fa /SYNTENY/genome4_chr.fa /SYNTENY/genome5_chr.fa /SYNTENY/genome6_chr.fa /SYNTENY/genome7_chr.fa /SYNTENY/genome8_chr.fa
        (command exited with non-zero exit code)
Shutting down, this might take some time.

Please advice how to troubleshoot.

Thanks
Abhijit

[lcoombe]

Hi Abhijit,

1. In that case, I would set -d to 25

2. Getting 'no paths found' in the error message like you saw is due to the algorithm failing to find any large regions of synteny with your input files parameters. I find this sometimes as I try to compare a higher number of genomes that have high nucleotide divergence - as you have there, judging by the result of Mash.

The first thing I would try is reducing the number of genomes that you are comparing to see when you get paths/synteny blocks generated. Like the message suggests, you can also try adjusting the k and w parameters - you could start by trying to reduce the k value (ex. to 18 or 20) and the w value (ex. to 500).

[sanyalab]

Hi Lauren,

Thank you for replying. ChatGPT suggested exactly what you mention. Although, it started hallucinating after about 100 lines 😃

I tried several kmers for my dataset. Optimum is k=19, w=500 for all the species. Is there a way to construct the synteny incrementally? there are species that are closer than others. I can create a quick tree of these species to get the relationship.

Thanks
Abhijit

[lcoombe]

Hi Abhijit,

I'm glad that ChatGPT agrees! 😃

Unfortunately, right now, we do not have a mechanism to combine results of separate synteny results together. It is in our future plans to introduce new approaches that will help the tool to be more robust to comparing a high number of more divergent genomes, though.

[github-actions]

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 ntSynt!