Publish panhuman index construction workflow

[bede]

Either here or Zenodo. cc @pmoris

[microbemarsh]

Hey @bede, related to this, would you be able to share the .fasta files for the panhuman and panmouse indices somewhere too please?

[bede]

Hey @microbemarsh, I've made an end to end snakemake workflow to do this which I'll either add to this repo or a new one. I'll share once tested.

[microbemarsh]

Great, thank you!

[bede]

Hey, thanks for patience, I've automated a workflow that reproduces the panhuman-1 index by downloading all of the genomes and then indexing them either in parallel (fast) or on a stream (lower memory) before combining and masking them.

https://github.com/bede/deacon-indexes

I'm in the process of a running a final validation, but have successfully reproduced a k-mer-exact version of panhuman-1 from scratch with this repo:

Note that the file checksums may not match due to k-mer reordering during parallel indexing, but you can easily verify that two indexes contain indentical k-mers using either deacon index union or deacon index diff.