I'm using the Novaseq 6000 and HiSeq 4000. Assembled alone, the Hiseq data has little missing data and many reads per sample, Novaseq has more missing data but still is useable. When I assemble them together, Hiseq individuals have few or no SNPs. I've checked trimming for both datasets and that does not appear to be the issue.

Assembling using ipyrad, I've assembled de novo and mapped to a reference.