Ok, I was slightly off: It was $45 (US) for 15 million paired end reads per sample, including lib prep, but excluding DNA extraction and labor, 192 samples pooled onto a NovaSeqX 3000M kit.

I believe 50-100M reads. Even though these are not well sampled environments, our pipeline does not rely on assembly or mapping and I believe we'll be able to get all the information we are looking for...

Anyway, at the same price and less work we get a nice metagenome of sufficient depth to get good community profiles as well as analysis of functional genes. No MAGs of course, but that is often not the objective.

This leads to higher costs, about 70 CAD per sample for amplicon sequencing perhaps including PCR and DNA extraction, but excluding the labor of manually extracting DNA and doing PCR, checking products with gels etc. That's the advantage of a large research conglomerate like Vienna I guess

Well that is really good value for your money! Here, we have problems making effective use of the higher sequencing capacity of the NextSeq instruments for amplicon sequencing as well as lacking effective robotics I guess. We have too few samples generated per month to reduce costs.

Just curious: what exactly do you get for those 12 euros and what is the turnaround?

That being said... good primers always welcome of course.

Perrsonally, I believe the days of amplicon sequencing are over. We now have #SingleM!

Personally I thought this paper was well done. AI has good use cases and CBC is using it transparently.