graph resource issue

[jiadong324]

Hi,

I found hprc_y1 graphs were also released here: https://zenodo.org/record/6983934#.Y4WLWOxByDU. What's the difference of the graphs given in this repository.

Thanks!

[glennhickey]

The graphs in this repo are the ones analyzed in this paper
https://www.biorxiv.org/content/10.1101/2022.07.09.499321v1

The graphs in your zenoodo link are more recent: they use a newer version of Minigraph and a slightly newer version of CHM13 (v2.0 instead of v1.1 -- I think the only difference here is the addition of a Y chromosome)

[jiadong324]

Are graphs in the repo (minigraph, minigraph-cactus, pggb) all contains 90 haplotypes? Since I want to check the paths for certain haplotypes, such as HG00733 in the graph created by minigraph. But it seems to me that the graph GFA format dose not contain line structure P. Thus, I am wondering how can I get the path for specific haplotype.

[glennhickey]

Minigraph stores path information in S-line tags. But it only contains the reference path in full, other samples just appear as insertion fragments (as each node is part of exactly one path described this way).

Minigraph-Cactus stores the haplotype paths as W lines, as these store metadata not available in the P-lines. You can convert them to P-lines with gfa-wp where metadata is embedded into the names.

PGGB should have the P-lines as-is.

[jiadong324]

Thanks for your explanations!