lower N50 after running 3D-DNA

[ziqing he]

Hi all,

I used hifiasm to assemble primary genome (estimated size 2.5G) with HiFi data and ONT utra long data. I got 125 contigs and N50 is 132,917,691. After runing 3D-DNA, I got 3137 scffolds and N50 is only 10,925,166.
The following is the code I run:

hifiasm -o hamONT.asm -t 30 --primary --ul pass.ul.fq.gz HiFi12.fastq.gz

bash /biosoft/juicer/CPU/juicer.sh \
-d /Hic_asm/hifiont \
-D /biosoft/juicer/CPU -y restriction_sites/hifiont_DpnII.txt \
-z references/hifiont.np3.fa -p restriction_sites/hifiont.chrom.sizes -s DpnII -t 25

run-asm-pipeline.sh -r 2 /Hic_asm/hifiont/references/${fasta} /Hic_asm/hifiont/aligned/merged_nodups.txt

The Hic heatmap of hifiont.0.hic

The hic heatmap of hifiont.final.hic

Any advice for me to improve the assembly? What parameters should I use?

Thanks in advance!