[pb-assembly] Script for generating haplotig placement files

[gbdias]

Hi,

• pb-assembly now outputs a very useful PAF file with the placement of haplotigs relative to the primary contigs after Unzipping.

New! Placement Files -haplotig placement (PAF format) generated in 3-unzip stage

• I assume this file is obtained with minimap2, but some processing is done to the output.
• Could you point me to the script that generates this file?

Thanks, Guilherme

[gbdias]

• The haplotig placement is not valid after Arrow polishing because the lengths of primary and haplotigs change a little bit.
• It would be super useful to have the haplotig placement file be generated again in the 4-polishing stage.
• I am trying to do this now and having a hard time with it.

[gbdias]

• I needed to obtain haplotig placement info to submit a phased assembly to genbank (haploid+alts).
• Repeats and structural variants will create a ton of weird alignments between haplotigs and primary contigs.
• However, by retaining the single longest alignment segment for each haplotig one can usually find the correct placement.
• I achieved this using minimap2 with the preset -x asm5.
• The alignment will sometimes not span the entirety of the haplotig, but the GenBank alternate locus placement format (https://www.ncbi.nlm.nih.gov/assembly/docs/submission/#ex4) allows one to specify the length of the alignment tails.
• It would be nice to have more precise boundaries for haplotig placement but this might be enough in some cases.

[zeeev]

Hi @gbdias,

I thought I would mention that FALCON-Phase, in its first stage, produces haplotype placements using mummer. Perhaps they might work better in this case?

[gbdias]

Hi @zeeev , thank you for the tip.

Apparently GenBank is no longer taking haplotig placement info into account when you submit phased assemblies, so I no longer need this placement info for that particular purpose. But FALCON-Phase placement could be useful in many scenarios.

Thanks