How many control samples?

[vfonti]

Hello! Provided that as many negative controls as better, how many many should I plan to have? Right now, I have about 91 samples and three negatives, where the latter are DNA extraction blanks that have been processed and eventually sequenced as they where true samples. For the next experiments I will certanly push to have more negative controls but how to decide how many?

[benjjneb]

In the paper introducing decontam, we discuss the number of negative controls in the discusison section on Experimental design: https://link.springer.com/article/10.1186/s40168-018-0605-2#Sec14 (and see also Fig S6).

We recommend at least 5, but more helps as some contaminants are "patchy" in the negative controls.