A simplified Pretext workflow specs

[nekrut]

A simplified workflow should take the following inputs:

• hap1
• hap2
• hap1 name
• hap2 name
• HiFi reads
• HiC F reads
• HiC R reads
• Telomere (units)

It generates a PretextMap snapshot and a dataset suitable for running Pretext View

[fubar2]

@nekrut @Delphine-L

I've been using this subworkflow derived from Delphine's version at https://vgp.usegalaxy.org/u/fubar/w/ebphib-pretext-hic-2 . It probably works fine as a workflow but I cannot get the parent EBPhib-big workflow to finish on vgp because of the samtools view OOM problem reported as a bug on 24 and 20th October. Team meeting time so probably lost somewhere.

If samtools view can have some more ram and I can get the whole WF to finish, could show you the end result JBrowse2 with the cools as HiC tracks.

The subWF creates pretextmap for viewing and cool tracks for JBrowse2 for each haplotype separately and both together.
Cannot get the coverage bigwig to show in Pretextview after adding it to the pretextmap so not sure telomeres will work either. They're available in JBrowse with the new HiC tracks anyway.

Also creates a paf containing all the contact pairs from the final merged bam, for the new and amazing interactive HiC Jupyter notebook viewer.

Top row is H1-H1 (call it cis for the haplotypes) then H2-H2 pairs and bottom row is H1-H2 trans pairs.

When first opened they show all the data - about 14 million pairs.
Clicking anywhere on a plot shows the coordinates in the row above.
The notebook normally produces larger images but I squeezed them down to fit into this family photo.
Here's mUroPar1 - the parka squirrel arima HiC data.

[fubar2]

T2T HG002 is running in https://usegalaxy.org/u/fubar/h/ebphib-pretext-noextra-hg002
It is very slow right now. Only a couple of trivial steps completed in the last 3 hours.
Is vgp.usegalaxy.org ok?

[fubar2]

https://github.com/fubar2/hv-notebooks

[Smeds]

I have an updated version of the pretext workflow Delphine worked on (galaxyproject/iwc#584)

[mvdbeek]

Can I move this issue over to the IWC ? I think we should mark this repo as deprecated

[fubar2]

@mvdbeek: which one do you want deprecated?

[mvdbeek]

This repository. This discussion should happen on the IWC.