Hi, guys. I have almost 300G paired-end (150bp) Illumina short reads and my server has 64 cores and 500G RAM. Can I use Abyss assemble it? The genome size is about 4G.
You can certainly try. How many reads is this? What is you insert size? Did you calculate the sequencing coverage?
It is certainly worth a try. 500GB RAM is in the right ballpark for a genome of that size.
The memory usage depends on the number of distinct k-mers in your sequencing data, which depends on genome size, sequencing coverage, and sequencing error rate.
You can reduce the memory requirement by adding the quality trimming option (e.g. "q=15") to your abyss-pe command. That will trim the tails of the reads and remove a lot of the error k-mers.
Thanks for you advise. I will try.