Statistics analysis of assembled genome from PacBio HiFi reads using Hifiasm

Question

[K]

Hi,

I had consensus reads from a PacBio sequencing. I used the assembler hifiasm to create an assembly on these CCS. I got five .gfa files. I have transformed .gfa file to .fasta contigs using bandage and awk command.

What can be used to get more statistics on the assembly?

[Dave Carlson]

Quast and BUSCO would be my initial suggestions, but it really depends on what specific stats you're looking to compute.

[K]

bandage gave me N50, total length but I am particularly looking for coverage and depth of the assembly.

[Dave Carlson]

To get that, you should maps the reads back to the assembly (e.g., with minimap2 for long reads) and then use a tool like samtools or mosdepth to get the depth from the sam/bam file.

Answer 1

[Billy Rowell]

You can get N50/NG50 from https://github.com/lh3/calN50.

Answer 2

[gconcepcion]

This py3 script will get you basic fasta stats: https://github.com/PacificBiosciences/pb-assembly/blob/master/scripts/get_asm_stats.py

[K]

this is really great. I am getting the same output as in bandage. However, how can I get the coverage and depth of the assembly?