How To Tweak The Priceti Assembly Package For Viral Assemblies

Question

[rst]

Does anyone have experience using the PriceTI assembly package? Specifically are there any parameters that people have found are useful in tweaking assemblies for the better? In this particular case it is a viral assembly from a complex pool of reads where the virus represents only a small fraction of the total reads. Thanks.

Answer 1

[ucpete]

I almost always use target mode for viral genome assemblies in metagenomic contexts. I'll typically leave target mode off for the first 3 or so cycles, then alternate 2 on, 2 off. It really helps. But it depends on how your assemblies are suffering. Target mode helps if PRICE slows down due to assembly of low complexity or unrelated regions -- in other words, it helps if you're overassembling. If you're struggling to get anything to assemble at all, or underassembling, you can tweak the minimum overlap parameters. Or if your data are error-prone (like 454, PacBio, or non Illumina), you may want to tweak the percent ID parameters. I've got a lot of experience and have had great success with PRICE, so don't hesitate to provide more detail and I'll try to respond. Also the author is really responsive so you could try emailing him if you can provide him sufficient detail to try to help troubleshoot.

[rosaless]

Hi, Any advice on using PriceTi for closing a virus genome in regions with long stretches of repetitive segments (~10kb)?

Thanks!