Separate haplotype-resloved untig in GFA about hifiasm HOT 1 CLOSED

Hi Zhigui,

Have you ever looked at the unitig graph (r_utg.gfa/p_utg.gfa) at Bandage? I guess for species with 1.8% het, most haplotypes have already been fully separated. The graph should like Figure 1 in this blog (https://medium.com/pacbio/hifi-assembler-series-part-1-hifiasm-a-fast-haplotype-resolved-genome-assembler-bd2f30dab571). If you just want to get all phased assemblies, r_utg.gfa/p_utg.gfa consists of all information.

However, although most chromosomes have been separated, we don't know which one is parental/maternal. So you still need trio information. We plan to release a new version with trio in the next 1 or 2 weeks, which can tell you the exact parental/maternal information of each unitig.

BTW, right now the p_ctg.gfa consists of one mixed primary assembly. But I guess it might be not useful for you. And to visualize graph by Bandage, I recommend you to utilize *noseq.gfa. These gfa files don't have sequences, while still keep the graph structure.

from hifiasm.