What methods/parameters would you recommend for running this on potentially eukaryotic MAGs?

[jolespin]

I noticed that EukHeist classifies MAGs as 90% EukRep hits as eukaryotic.

If you're going with this approach, a few questions:

1. What parameters would you use?
2. Would you weight the contigs based on size?
3. Would you split up the contigs into windows?