search

chhylp123 / hifiasm

Hifiasm: a haplotype-resolved assembler for accurate Hifi reads
MIT License
528 stars 86 forks source link

hap1 and hap2 generated from different versions are of different sizes #463

Open xzhoubayer opened 1 year ago

xzhoubayer commented 1 year ago

The crop genome is inbred diploid. Estimate genome size is 1.2 Gb.

v0.15 output: primary--1.12Gb, hap1-- 1.03Gb, hap2--1.02Gb; v0.19 output: primary--1.1Gb, hap1--831Mb, hap2--470Mb. Default parameter settings were used.

Could you please give some suggestions which version should be used? what is the best parameter setting should be used?

chhylp123 commented 1 year ago

See FAQ here: https://hifiasm.readthedocs.io/en/latest/faq.html#how-can-i-tweak-parameters-to-improve-hi-c-integrated-assembly. This works for both Hi-C and dual assemblies.

ericgonzalezs commented 6 months ago

I am having the same problem. An old version gave me the size I was expecting (I am sorry I don't know the version, probably a version from 2022 ), and the latest version 0.19.8-r603 gave me very different sizes for hap1 and hap2, using default options with the same Hifi and Hi-C data. Is this expected in the new versions? Does the -s default parameter was changed? What do you think can be the biggest difference between old and new versions?

shehongbing commented 2 months ago

I also have the same problem. My hifiasm is 0.19.9-r616. I combined the HiC, HiFI and ONT data, and the hap1 size is ~300 Mb, hap2 size is ~ 960 Mb (consistent with my expected genome size).