Entering edit mode
5.1 years ago
lovenlong
▴
10
Hi everyone,
I'm trying to assemble a plant genome (~400 Mb) with ONT data (SumBase ~20 Gb, 36.5k reads, mean len=53k). Due to some reasons, I have to do the assembly with my Desktop (i5-7400, 16 GB RAM, 2 TB disk).
I did overlap mapping with minimap2:
minimap2 -x ava-ont -k15 -Xw5 -m100 -g10000 -r2000 -t 2 -I 2G --max-chain-skip 25 ONT.fastq ONT.fastq > ONT.ovlap.paf
after ~70 hours, it finished and generated a PAF file with 689 GB in size.
then I tried to get gfa with miniasm (miniasm -f ONT.fastq ONT.ovlap.paf > ONT.gfa), but it failed due to core dump (Seg fault).
my questions are:
- is it ok for the very large PAF resulted from minimap2 ?
- how to calculate the PAF file size with a certain reads dataset ?
- is possible to assemble my data just using 16 GB or 32 GB RAM ?
thanks
I'd suggest you take a look at wtdbg2.
It works, thank you.
Canu is also very good for long read assemblies