Hello SEQanswers community! I want to sequence bacterial, fungal, plant and arthropod libraries on one MiSeq run from the same faecal samples. I know output can be biased towards shorter reads when multi-marker sequencing, but what is the largest difference in length that can be sequenced together without much difference in output? I'm considering 4 sets of primers that will give amplicons of approx 353bp, 400bp, 406bp and 465bp (not including adapter, index, pad & linker). Would the difference between 353 and 465bp be too great?

For example, da Silva et al (2020) had an average output of 11k reads/primer/sample for 18S, 12k for trnL, 10k for 16S and 6.7k for COI from 93 bird guano samples on a single MiSeq run (v2 250PE). I thought I might get around this by not pooling by sample, eg only having 96 samples but 96*4 libraries, one for each marker. Would this be overkill? Should I just save $ and pool by sample?