Hi all,
I’m about to start an error analysis, and I was wondering how to gauge the minimum number of sequences to include in the mock community. Also, is it safe to assume that the relative abundances of mock community members will have an impact in the seq.error analysis? If so, should mock community structure resemble that which we expect to find in our sequenced community, at least in terms of abundant taxa?
Many thanks.
Pedro.
Pedro,
The answer to each of these questions really is that It depends on what you’re trying to do…
I was wondering how to gauge the minimum number of sequences to include in the mock community
Some have resequenced an individual genome’s 16S fragment (E. coli) and other have put in dozens of sequences. The latter actually have a hard time making sure that all of the sequences make it into the pool of DNA. It’s also a good idea to put things in that are more closely and distantly related to each other.
is it safe to assume that the relative abundances of mock community members will have an impact in the seq.error analysis?
That sounds like a testable hypothesis. I could imagine it affecting things like chimera formation. Also, some populations may have higher sequencing error rates than others.
should mock community structure resemble that which we expect to find in our sequenced community, at least in terms of abundant taxa?
This is what we try to do to some extent. Our mocks come from sequences we expect to see in the gut, because that’s what we sample. Some people use clones to include uncultured fragments, some use genomic DNA from cultured bugs. Some try to keep everything at an even abundance (and typically fail), others try to stagger things.
Ultimately, the composition of your mock depends on what you’re trying to do. Do you just want a control for your pipeline? Do you want to look at biases? Chimera formation?
Hello Pat,
Thank you for your reply. The mock community would be used for pipeline validation (and hopefully this will also reveal chimera formation biases).
Thanks,
Pedro.