Using juicebox assembly tools after a scaffolding tool that is not 3d-DNA

[Dustin Sokolowski]

Dear Neva and Olga,

Short question: 

Can you edit an assembly file such that the "chr" can be equivalent to the "scaf"?

Long reason and message:

Thank you for your amazing tools! I am assembling a de novo diploid genome using ONT nanopore long-read data, Hi-C data, and 10X linked reads. 10X linked reads gives me the additional advantage of leveraging some long-read information that our Hi-C data may have missed for technical or biological reasons. My planned pipeline is:

Trio Bining with TrioCanu > align with flye > polish with racon/medaka > scaffold with Juicer-3dDNA > Extra scaffolding with scaff10X > Clean up assembly with JBAT.

I remade the .assembly file by re-running juicer-3DNA and the scaffolded genome using the above pipeline. QUAST and BUSCO suggest that re-running juicer-3dDNA after the above pipeline decreases genome quality (I'm happy to send the information if you're interested), so I was hoping to use JBAT with the assembly.0.assembly file instead of the assembly.final.assembly file. This being said, the assembly.0.assembly file has 1  superscaffold still, and I've found that after 2 hours of clicking I can't get the "chr" blue boxes around the "scaf" green assemblies, making it so I can't fix the (roughly) half dozen or so genomes that are split in half (see attached). With this in mind, can the assembly files be edited such that I can start with the "scaf" to make the handful of needed changes.

Thanks so much!

Dustin

 

[Olga Dudchenko]

Hi Dustin,

You do not need to rerun 3D-DNA to generate an assembly file and hic. Look into visualizing the draft (i.e. any fasta file) in the genome assembly cookbook. This will give you an idea of how to generate what you need, and you can review that in JBAT. You'll still need 3D-DNA tools to generate the fasta based on your review.assembly file. 

Best,

Olga