I have a PacBio HiFi assembly of 1.1 Gb from a heterozygous species. I have aligned this assembly against a reference genome which is around 0.9 Gb. I can see that there are quite a few INDELs, Duplications, Inverted regions in my Query sequence on the Y-axis (PacBio Hifi) compared to the reference genome.

I want to check if these structural variations are real or only resulting from sequencing errors or not. So, I want to manually curate the PacBio assembly. Because I can see at which contig the structural variations are.

Can you please suggest any tools for manual curation of a genome assembly to investigate whether these structural variations are sequencing errors or real sequence variation?

I have uploaded the dot-plot below. Thank you.enter image description here

  • $\begingroup$ based on your other question, I would suggest mapping the hi-c reads to this assembly and manually polishing in juicebox $\endgroup$ Commented Jan 27, 2022 at 20:46
  • 1
    $\begingroup$ So, you mean take the HiFi fasta and map the Hi-C fasta on it? Or do I need raw reads from the Hi-C reads? $\endgroup$
    – Anik Dutta
    Commented Jan 28, 2022 at 9:16
  • $\begingroup$ You would need to map the raw Hi-C reads. the FASTA is the outcome of a lot of procedures that remove information. It is the Hi-C reads have the actual proximity information about which regions are close to each other on the chromosome. You can use that to correct structural errors (if any). If those are truly errors then they will look weird on the Hi-C heatmap and you can correct them manually in juicebox: aidenlab.gitbook.io/juicebox/assembly-tools $\endgroup$ Commented Jan 28, 2022 at 17:41
  • 1
    $\begingroup$ Ah ok. Thanks for clarifying the confusion. It is clear now. $\endgroup$
    – Anik Dutta
    Commented Jan 28, 2022 at 18:16


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.