Write to Fasta slows down in 25000X coverage range

[AyushSaxena]

We are trying to simulate lots of reads for a small plasmid. Achieving 100k coverage is common, and that is the range I was trying to simulate. I noticed that the tool can very quickly simulate upto 20k-25k reads, but slows down dramatically after that (barely adding 100 reads in a minute). Maybe I'm using the tool wrongly, but I thought I should bring it to your attention.

Thank you
Ayush

[ndierckx]

Hi,

I never tried that high coverage, I will have a look at it. How large is the plasmid, then I will try to replicate your run?

[AyushSaxena]

Hi,

The plasmid is ~6000bp long. The only change I've made to the source code when I'm running it is that I've installed the dependencies in a local lib directory and the perl script reads the dependencies from that directory. Everything else is the default. I'm using CCS error profiles.

20k runs really fast, and if I recall correctly, I started seeing the slowness when I requested 25k, but you could just try 30k perhaps to test it.

Thank you
Ayush

[ndierckx]

Hi,

You are right, time seems to go up a lot with very high coverages.
This is because it tries to get unique sequencing lengths for every read. I will change this to make it relative to the amount of simulated reads. I will upload a new version tomorrow that will be as fast as for low coverages.