Deduplication and alleles frequency estimation with strain stracking pipelines about midas HOT 4 CLOSED

Hi Camille,

Thanks for your question. I agree that deduplicating reads could be a problem for accurately quantifying allele frequencies, which is why I would recommend not deduplicating. That's also the recommendation from this paper.

Thanks,
Stephen

from midas.

Great, thanks for your reply and the very useful reference.
So, may I ask you, in your MIDAS paper, you used metagenomics data from the HMP and did a diversity analysis on those using MIDAS.

Did you used already processed data (with deduplication) provided on their repository (hmpdacc) or decided to use/could access raw data to process their your own way?

Many thanks,
Camille

from midas.

For gut metagenomes (used mainly as a control dataset for strain tracking) I downloaded datasets directly from the SRA.

For other body sites (used to assess the % of microbes from unknown species) I used datasets from the HMPDACC.

Thanks,
Stephen

from midas.

Ok, thanks!

Best,
Camille

from midas.