The chromosome-length genome assemblies presented here were created following the strategy outlined in detail in our Genome Assembly Cookbook.
In brief, in situ Hi-C data was aligned to a draft genome assembly using the Juicer pipeline from (Durand, Shamim et al., Cell Systems, 2016).
Many different types of draft assemblies were used as the starting point to create the chromosome-length references. When available, drafts were downloaded from NCBI: the links can be found on the individual assembly pages.
In several cases the draft genome assemblies were shared with us by collaborators. These were based on a variety of different assembly strategies: please refer to the links listed on individual assembly pages for details.
For the remaining assemblies $1000 de novo strategy was employed. The initial contig set for these assemblies was generated using w2rap contigger from (Clavijo et al., Genome Res., 2017).
The 3D-DNA pipeline from (Dudchenko et al., Science, 2017) was used to error-correct, anchor, order and orient the pieces in the draft assembly. The 3D-DNA visualization module, in conjunction with Juicer Tools, was used to create contact maps for the draft and the final genome assemblies.
As a rule, 3D-DNA was run without parameter tuning. Instead, a manual review step using Juicebox Assembly Tools aka JBAT (Durand, Robinson et al., Cell Systems, 2016; Dudchenko et al., bioRxiv, 2018) was employed to polish the genome assembly output by 3D-DNA.
Check out this tutorial and barrel clover assembly video to get an idea of how JBAT works. Follow this link to download the .hic and .assembly files and follow along with the tutorial yourself!
The Hi-C contact maps available on individual genome assembly webpages were embedded using the Juicebox.js software from (Robinson et al., Cell Systems, 2017).
If you have any questions about our tools, do not hesitate to write on our forum!