During this process we found (what we believe) to be primer contamination within our N1 gene primer set. During our initial pilot sequencing run, we observed a high concentration of reads mapping to our mutated synthetic spike, however we had not built or included the synthetic spike in this run.

After a couple days of RT-QPCR diagnostics, we found that our library prep primers were all contaminated with oligo for our synthetic spike-in construction (that was ordered on the same day) and we could create both RT-PCR and PCR product without adding template RNA or DNA. (Diagnostic Experiment here)

Here we can see the N1 primer set can amplify SARS-CoV-2 sequence in the absence of added template (in a human cell line lysate background), similar to a positive control.

We should have foresaw this, but as depicted below our forward primer for construction of the T7 template of the N1 spike has 12 nt of homology for our N1 assay's RT primer and participates as a DNA template of reverse transcription

We diagnosed this by performing both a gradient on the RT step and the annealing step of the PCR and finding the amplification was only inhibited differently from a postive control in the RT annealing gradient (data not shown).

In conclusion, our N1 datasets are obscured by this primer issue IDT claims to have sorted this issue out and in general flags sequences with homology for CDC assay primers but in this case the homology was too short to be flagged. Luckily our S2 primers didn't have this same problem, and it is probably why the data look better overall.