Hello, I used parental short reads to phased HIFI reads, and the N5O of the phased haplotigs is 12, 246, 342 and 12, 727, 763 respectively. I would like to ask if I can improve contiguity if I add HiC reads again. Additionally, does the hic data have to come from the same individual as the Hifi data to perform genome phasing?
The following is the code I have run.
hifiasm -o ha.asm -t 20 -1 /Hifiasm/pat.yak -2 /Hifiasm/mat.yak /Hifiasm/HiFi.fastq.gz
Here is the code I want to run
hifiasm -o ha.asm -t 20 --dual-scaf -1 /Hifiasm/pat.yak -2 /Hifiasm/mat.yak --h1 read1.fq.gz --h2 read2.fq.gz /Hifiasm/HiFi.fastq.gz
Hello, I used parental short reads to phased HIFI reads, and the N5O of the phased haplotigs is 12, 246, 342 and 12, 727, 763 respectively. I would like to ask if I can improve contiguity if I add HiC reads again. Additionally, does the hic data have to come from the same individual as the Hifi data to perform genome phasing?
The following is the code I have run.
hifiasm -o ha.asm -t 20 -1 /Hifiasm/pat.yak -2 /Hifiasm/mat.yak /Hifiasm/HiFi.fastq.gz
Here is the code I want to run
hifiasm -o ha.asm -t 20 --dual-scaf -1 /Hifiasm/pat.yak -2 /Hifiasm/mat.yak --h1 read1.fq.gz --h2 read2.fq.gz /Hifiasm/HiFi.fastq.gz