Ability to output DADA2 metrics

Ideas and Suggestions
1

DADA2 has the ability to provide read counts across samples for each stage of processing. It would be very nice if QIIME2 could pass this information on. It’s a big time saver in identifying what went wrong when read counts are lower than expected (a question that has popped up multiple times here). Even when the counts look reasonable, I sometimes get asked for this information by a PI, which I currently can’t do without rerunning DADA2 outside of QIIME2.

The command for this is in the DAD2 tutuorial -https://benjjneb.github.io/dada2/tutorial.html

getN <- function(x) sum(getUniques(x))
track <- cbind(out, sapply(dadaFs, getN), sapply(mergers, getN), rowSums(seqtab), 
rowSums(seqtab.nochim))
# If processing a single sample, remove the sapply calls: e.g. replace sapply(dadaFs, getN) with getN(dadaFs)
colnames(track) <- c("input", "filtered", "denoised", "merged", "tabled", "nonchim")
rownames(track) <- sample.names
head(track)

Since we are a core facility that processes multiple projects from the same miseqs, it would also be nice to have the error profile plots for forward and reverse reads

plotErrors(errF, nominalQ=TRUE)

Thanks!
Chris

1 Like
3

Hey @Chris_Hemmerich!

You’re on the same wavelength as us, so to speak! We’ve already got an issue tracking this. I don’t know when it’ll happen for sure, but hopefully soon.

Thanks!

1 Like
5

QIIME 2 2017.12 is out now, and it includes more detailed logging when running DADA2!

6

I tried it out, and thanks - this is great!

2 Likes