In the diploid assembly, hfiiasm identified a value that did not exist in the k-mer plot as the "homozygous read coverage threshold".

[xujialupaoli]

Thank you very much for providing such useful assembly software！ The hifiasm version I am using is 0.19.8-r603 I am trying to assemble a diploid genome, first using the following code to assemble, the size of the output 2 haplotype genome is 84M. hifiasm -o Dohr1.asm -t 24 hifi_2haps.fq I'm very surprised, my hifi data quality is very good, but there is only one peak in the kmer plot. The corresponding value of this peak is 82. But hifiasn automatically identifies the homozygous read coverage threshold as 164. I don't understand how this value is obtained? hifiasm_log_himozygous_auto.txt

[M::purge_dups] homozygous read coverage threshold: 164

Afterwards, when I set --hom-cov to 82 according to the usage method of Hifiasm, the output result was two haplotype assembly results of 168M. hifiasm -o Dohr1.asm -t 24 --hom-cov 82 hifi_2haps.fq

hifiasm_log_himozygous_82.txt

Could you please help me analyze which output result is more reliable? Why does hfiiasm in the diploid assembly identify a value as "homozygous read coverage threshold" that does not exist in the k-mer plot? Why does my kmer plot only have one peak when the data quality is very good?

Very much looking forward to your reply！