search

lbcb-sci / herro

HERRO is a highly-accurate, haplotype-aware, deep-learning tool for error correction of Nanopore R10.4.1 or R9.4.1 reads (read length of >= 10 kbps is recommended).
Other
136 stars 9 forks source link

hifiasm options #29

Open chklopp opened 1 month ago

chklopp commented 1 month ago

I've tried herro on one of our read sets (with R9.4.1 model) and when I compare the reads before and after correction, by aligning them on the reference assembly, the result is very convincing (very few INDEL errors left.) But when I assemble the reads with hifiasm 0.19.8 I lose close to all reads (95%) during the first hifiasm correction step : comparing first and second kmer histograms in hifiasm logs. This is not the case when I assemble HiFi reads.

Which hifiasm options should I change to limit this phenomenon?

chklopp commented 1 month ago

After after a second check the drop in kmer coverage is also found with HiFi reads. Still I do not understand why the assembly metrics are so low : hifiasm is using a very low number of reads for the assembly, this can be seen in the gfa coverage values

h1tg000001l 114415 6 h1tg000002l 1935040 3 h1tg000003l 485308 3 h1tg000004l 113763 0 h1tg000005l 54120 0 h1tg000006l 82359 0 h1tg000007l 3376377 2 h1tg000008l 505683 2 h1tg000009l 1826044 2 h1tg000010l 4045620 2 h1tg000011l 151854 1 h1tg000012l 172642 0 h1tg000013l 75530 0